Changelog

Claude, ChatGPT and other agents connected to VidNext — and the API — now reach the rest of your account, not only new videos.

• All your videos. Find any video — also the ones you made in the app — by words from its title, prompt or channel, and read or change its YouTube title, description and tags.
• Thumbnails after the video. See a video's thumbnails (your agent can look at them), make new ones, remake one, change one by an instruction like "make the title bigger", and choose the main one.
• Channels and competitors. Add and edit channels and their profiles, analyze a YouTube channel, find competitors with AI, and see which of their videos took off.
• Ideas and the calendar. Generate ideas, approve the good ones, add your own, put them on the calendar, and make a video from an idea — like Generate video in the Planner.
• The Planner. Ask what's posting today and what still needs a video, what to schedule (and have it put on the free slots), which videos to start next, and how a channel is doing — its health, runway and YouTube numbers. Move, skip or add posting slots and mark videos posted.
• Thumb Lab and styles. Make thumbnails from a brief in Thumb Lab, edit them, save the one you like as a style, and set a channel's default styles.
• Ask before spending. Choose for each connected app and key whether it asks you first: before anything that spends credits, it tells you what it's about to do and the estimated cost, and waits for your OK. Apps like Claude Code show VidNext's own question; in others your AI asks you in the chat. Apps you connect by signing in ask first unless you change it; keys don't. Either way it's charged as usual — see Before it spends credits.
• Everything costs what it costs in the app, and apps like Claude ask you before anything that changes or deletes something. See The Planner, Channels, competitors and ideas, Thumbnails and Thumb Lab and the tool list.

A scene about a real person no longer gets a generated picture of them, and finds real photos of them more often.

• Never an invented face. Some picture slots of comparisons, before/after graphics and title designs could get a generated picture of the real person they showed. Every picture slot is now checked for a real person, brand or app first. When no real photo fits, the scene shows its design without the picture, or its narration as on-screen text.
• More real photos of people. A photo search for a person now also searches their name on its own and on Wikimedia Commons. A genuine photo of the person from another event or year is accepted when no photo of the exact moment is found.

A video no longer fails on its last step because one of its own files stopped loading.

• Failed on Final Assembly. Rarely, one of a video's own files, usually part of its narration, stopped loading from our storage for up to an hour while every other file loaded normally. The last step kept asking for that same file until it gave up. Now the video saves that file again as a fresh copy, checks that the copy loads, and continues with it. It takes a few minutes longer instead of failing. If it still can't finish, the video stops without charging you, and Resume on the project page continues it.

Our team can now give your account a personal discount — a percent off your first plan payment, your next plan payment, or every payment.

• Every plan, every cycle. A personal discount works on any plan, monthly or yearly, Enterprise included; one on every payment covers your top-ups too.

• Whole dollars. Prices under it are rounded down in your favor — 30% off Creator is $104 a month, and 30% off Pro yearly is $3,012 instead of $4,308.

• Where you'll see it. Signed in, the pricing page shows your price on each plan with the discount's terms under it; the plan card on the Credits page shows it too — on a plan, as the price of your next payment — and the top-up slider shows the discounted total.

• The fine print. It never stacks with another discount: each payment takes the biggest one. A next-payment discount is used by that one payment (a renewal or the charge for moving up a plan); one on every payment follows you when you change plans and stays until it is ended or its last day passes.

• In the docs. See Pricing & plans → Choosing a plan.

Add VidNext to Claude, ChatGPT, Claude Code, Cursor or VS Code with its address and your VidNext sign-in — no key to copy.

• Sign in with VidNext. Paste https://api.vidnext.ai/mcp into the app — in Claude: Customize → Connectors → Add custom connector — then sign in to VidNext and press Allow. It works in the Claude desktop app, on claude.ai and in the phone app.
• You see who's asking. The VidNext page shows the app, where it sends you back and what it will be able to do. You can give it a daily video limit right there.
• Connected apps. Settings → VidNext API lists every app you connected, each with its own Videos a day limit and Disconnect.
• The guide has the steps for every app. n8n and your own code keep using API keys.

A video no longer stops before Scene Assembly when a server restarts at the wrong moment.

• Stuck after the pictures were made. Rarely, a server restarted just as it was putting a video's finished pictures together, and the video then waited without moving on. Now that work is picked up again and the video continues by itself.
• Fewer restarts at busy times. Video Search now slows down for a moment instead of overloading a server, so restarts like this are rarer. At peak times a video's search can take a little longer.

Cash payouts for affiliates now go through the payout methods we currently offer — PayPal for now.

• One clear choice. The application's last step and the Payouts tab of the Affiliate program page list only the methods on offer. With a single method there is nothing to pick: the page simply says how you are paid.
• If your method is paused. A partner whose saved method is no longer offered sees a notice on the Payouts tab and gets one email asking them to choose another. Nothing is lost — cash you already requested waits and goes out with the first payout after you save a method that is on offer.
• Plain words everywhere. The hub, the claim confirmation and the public affiliate page name the methods on offer instead of a fixed list.

The Terms of Service (section 11.3) now say a payout account is one of the payout methods the partner hub offers. Details in Affiliate program.

• Replace a key. Replace on a key in Settings → VidNext API gives you a new key with the same name, permissions and limits. The old one stops right away — when a key leaked — or in 1 hour, 24 hours or 7 days, so your apps can switch first; its row shows when it stops. Here is what to do if a key leaks, and if a webhook secret leaks.
• No daily limit. API and MCP videos no longer stop at 100 a day — your balance decides. To cap an agent, give its key a Videos a day limit.
• Connect your agent, step by step. A new guide with pictures for Claude Code, Claude Desktop, Cursor, VS Code and n8n: where to paste the setup, how to check it's connected, and what to say to your agent. The Connect card links to it.
• Let your AI set it up. Copy instructions for your AI in the Connect card copies a short text for your assistant. Paste it into Claude Code and it connects VidNext by itself and checks that it works; in Claude Desktop, ChatGPT or another chat, the AI walks you through it. Your key isn't in the text — the AI asks for it. The same text is in the guide.

Every page of these docs now has a Copy page for AI button. It copies the page as plain text — its steps, tabs, notes and links — ready to paste into Claude, ChatGPT or another assistant, so it can answer your questions from the page or follow the steps with you.

A video no longer waits for hours on its last step.

• Stuck on Final Assembly. Rarely, one of the video's files stopped arriving halfway while the last step fetched it, and the video kept waiting for the rest without ever stopping. Now a file that stops arriving is fetched again within about two minutes, and a last step that stops making progress restarts by itself. If it still can't finish, the video stops without charging you, and Resume on the project page continues it.

A video whose narration contains an emoji, a flag or a rare character now finishes, and those characters show whole in every design.

• No more failed final render. When a scene's last words were typed on screen over a still frame and contained an emoji or a flag (for example a channel name like "Alerta Lujo 🇪🇸" in the call to action), the video stopped at the final render with an error. The typed text now handles every character, and the video finishes.
• Whole characters in every design. Typed titles, lists, quotes, captions, charts and name cards used to cut an emoji, a flag, a family emoji or a rare kanji (𠮷) in half for a moment, or showed it as an empty box. They now type, count and shorten text one whole character at a time.
• A failing design no longer stops the video. If one design can't be drawn while your video is being made, the video still finishes: a text design shows the same words as plain still text, and any other design leaves the scene's own footage or picture on screen. In the editor, such a design still shows a red error card, so you can see it and replace it.

If a video stopped at the final render this way, open it and choose Smart Resume in the Resume/Redo menu (see Redo & regenerate): only the final render runs again.

Videos find real photos again, a picture you add stays on screen for its whole sentence, a quiet voice is no longer drowned by the music, and long Planner prompts arrive in full.

• Real photos again. From October 20 one of the sources picture search relies on stopped answering, and many videos showed a generated picture where a real photo belonged. Picture search now draws on other sources and finds real photos again, and a picture it has already turned down for a scene is not looked at twice, so scenes get their picture sooner.
• Your own pictures keep their whole sentence. A picture you add to a video now stays on screen until the sentence about it ends. When a short paragraph is about that picture alone, the picture holds the whole paragraph, from its first word to its last. Before, a sentence could start on your picture and finish on a different one, or the line that names your product could play over another shot.
• The voice stays above the music. Some voices, in some languages, came out much quieter than others, so the music and sound effects covered the narration. Every narration is now brought to the same level before it is mixed.
• Long Planner prompts arrive in full. One-click videos from the Channel Planner used to cut a long prompt at 9,000 characters, so the end of the brief never reached the video. A planner prompt can now be up to 37,000 characters; the editor shows a counter, and a text over the limit is never cut: you shorten it before you save.
• Repeats you ask for are kept. When your prompt asks the narration to say something more than once ("repeat the key figure three times"), the script check no longer treats it as a repetition to remove.

These changes apply to videos created from now on.

You can now make VidNext videos from your own code, automations and AI agents.

• REST API at https://api.vidnext.ai/v1: create a video from a prompt or your own narration, with a preset for the settings and per-video changes — voice (cloned voices included), theme with its mode and Pro or Mini, thumbnails, captions, music, length, call to action and channel. Videos wait in your account's line and start as slots free up; results come by webhook or on request, with the video, its thumbnails and its YouTube text.
• MCP server at https://api.vidnext.ai/mcp with the same keys: connect Claude Code, Claude Desktop, Cursor, VS Code or n8n and just ask your agent for the video.
• Settings → VidNext API: keys with their own permissions and limits, one key for a whole team, webhooks, your presets' IDs, and a request builder that prices a request and copies it as curl, JSON or a prompt for your agent.

The API comes with the Studio and Enterprise plans; on another plan, write to support to turn it on for your account. See the VidNext API docs.

New to the Channel Planner? A How the Planner works button now sits next to the Channel Planner heading on every Planner tab. It plays a short video tour of the Planner. The button appears once the tour video is available.

Download now hands the video to your browser at once, so the file starts saving with your browser's own progress bar.

• Before, the Download button showed a spinner while the whole video loaded in the page, and only then saved it. A long video could keep you waiting with no progress shown.
• Now the browser saves the file directly, under the same name as before. This works for Download on the project page and the dashboard, the versions in the arrow menu, and Download Latest in the editor.
• Team members and download-only accounts download the same way.

See Exporting.

A video no longer stays on Download Clips for hours, and a long stretch of narration no longer shows one picture for minutes.

• Stuck on Download Clips. Rarely, saving a downloaded clip never got an answer, and the video kept waiting for it without ever stopping. Now a save that gets no answer is tried again within about two minutes, and a step that stops making progress restarts by itself. If it still can't finish, the video stops without charging you, and Resume on the project page continues it.
• One picture for minutes. When one part of a long script was far longer than planned, its narration could end up in a single scene, which then held the same picture for minutes. Long parts are now planned in shorter pieces, and narration that is too long for its scene gets scenes of its own, each with its own footage.
• A video that showed one picture for minutes can be planned again from the Resume/Redo menu. See Redo & regenerate steps.

When our team adds credits to your account, your credit history now says so — and they can come with a trial.

• A clear label. Credits our team adds show as Gift from the VidNext team in your credit history and in the credits menu in the top bar. They arrive at once and never expire.
• A trial when you need one. If your account was made after plans launched and has no plan yet, the credits can come with a trial: you can create videos, thumbnails and avatars without a plan until the date on your plan card, like a promo code's trial. See Credits from the VidNext team.
• The usual reminders. Two days before such a trial ends, and once it has ended, you get the same emails as with a promo code's trial. They follow your Credit balance setting in Settings → Notifications.

Scripts now sound like a person telling you a true story, and you choose how that person talks from a new list of writing styles.

• New writing styles. Twelve formats: Documentary (the default), Explainer, News report, Opinion & commentary, Story & biography, Mystery & true crime, Dark stories, Countdown, Sports story, High stakes, Coach and Ad / Promo. There are also five presenter hosts made for videos with an AI avatar: The Investigator, The Myth-buster, The Keeper of Secrets, The Insider and The Guide. Each style opens and ends its own way, and the picker shows a line of how each one sounds.
• Plainer, livelier scripts. The writer researches more real cases, myths and open questions, then retells them in everyday spoken words. It talks to the viewer, raises one question early and answers it by the end, and keeps numbers to the ones that matter.
• No content type step. News report always writes news, leading with what happened and who said what. Every other style works out from your prompt whether the topic is news or timeless.
• Calls to action in the narrator's voice. With the call-to-action box left empty, the narrator places the asks itself instead of using the same template line in every video.
• Long videos first. Shorts keep the current writer for now and use the closest of the earlier styles.
• Your presets keep working. Saved styles open under their new names: Professional narrator is now Documentary, and Storyteller is Story & biography. A preset saved with Sarcastic / Dry Wit now uses The Myth-buster.

See Scripts & prompting.

A long video no longer fails in its Video Plan step when one part of the plan comes back incomplete.

• What changed. VidNext plans a long video part by part. When the plan for one part came back cut off before its end, the step asked again in exactly the same way, got the same cut-off answer, and after a few hours the video stopped with an error, even when an earlier answer for that part was complete. Now the step asks again in a way that leaves room for the whole answer, and if the new answers are still incomplete, it uses the complete one it already has.
• What you may notice. In a part planned this way, a few scenes can stay on screen a little longer than the plan first gave them, because their narration is longer.
• A video that stopped this way can be continued from the Resume/Redo menu. See Redo & regenerate steps.

When you regenerate the script while it waits for your review, the new script now follows everything you chose for the video.

• What changed. Regenerating at Awaiting Script Approval used to write the new script with default settings: it dropped your writing style (a custom style included), your call-to-action choice and instructions, the language and the video format. Now the new script is written with the same settings as the first one, and it is checked and proofread the same way.
• Also fixed. Approving a script that is not in English now proofreads it in its own language. A custom writing style is kept when a script is revised after its quality check, and a video whose type was detected automatically keeps that type after you approve the script. Citation marks such as [1] or a source name in brackets that a script picked up from its research are removed before the voice reads it, and so are asterisks around a word or a title. And if the final polish of a finished script cannot run, you now get the checked script instead of a failed video.

See Reviewing before production.

Scene transitions got an overhaul: 13 new looks, the classics rebuilt, motion previews to choose from, and sounds that hit with the picture.

• 13 new transitions. Push Cut, Iris, Spin Zoom, Cross Zoom, Blur Dissolve, Crosswarp, Ripple, Burn, Light Leak, Pixel Dissolve, Shine Sweep, TV Off and Page Turn — 30 styles in all. See Transitions.
• The classics, rebuilt. Luma Fade now dissolves smoothly through the highlights of the next shot instead of popping in as flat patches. Every transition starts moving right at the cut and lands cleanly — no thin strip of the old shot left at the end — and the outgoing scene keeps playing through the transition instead of freezing.
• Transitions live in Design now. They sit right under the style packs: a pack brings its own set and the block shows a preview of each — hover one to watch it move. Customize opens every style with a moving preview — hover a card to play it, the speaker button plays it with its sound. Changing them keeps the pack's colours and fonts: the block says Custom set and offers Reset to the pack's set.
• A fresh set per style pack. Each pack now uses the transitions that suit it, and every transition belongs to at least one pack — see the table in Style packs.
• Whooshes on the beat of the picture. A transition sound now plays with its loudest moment exactly where the transition peaks, instead of starting early and peaking somewhere near it. Every look now has a short, soft sound made for documentaries, news and explainers — a light paper slide, a camera shutter, a soft air swish — that sits quietly under the voice, in landscape and Shorts alike.
• In the editor. Pick a transition from motion previews; changing its style or length keeps it centred on the cut. A transition's whoosh follows it — move the transition and the sound moves along, change its style and it gets that style's sound, delete it and the sound goes too. Adding a transition from the library now uses the style you picked (it always added Fade to Black) and snaps it onto the nearest cut.
• Music videos. Push Cut, Spin Zoom and Light Leak join the Music video set, and None now really gives a music video hard cuts only.
• Fixes. A transition after a scene made of several clips now continues the clip that was on screen at the cut (it could jump to another one). Captions and counters of an outgoing card (an image stack's "03 / 03", its photo caption) no longer stay on screen over the next scene while the transition runs. A cut inside one continuous AI presenter take stays a hard cut in the editor too, as in the video. The transition into a video's closing shot uses the calmest style you allowed instead of always a dissolve.

Saving a preset now keeps the logo you set on the create page.

• Everything in the Logo panel. The preset remembers whether the logo is on, which logo — your channel's, or one you uploaded just for the video — and the corner, size, opacity and fade-in you changed. Load the preset and the Logo pill is back the way you saved it, whichever channel you pick.
• Untouched settings keep following the channel. If you didn't switch the logo on or off, the preset follows each channel's Show the logo in new videos; a corner or size you didn't change stays the channel's own.
• One-click and Autopilot too. Videos started from the Channel Planner's Start now or by Autopilot with the preset get the same logo.
• Teams. A team preset brings its logo with it. Your copy of a team preset keeps the owner's uploaded logo while presets are shared, and says Logo not shared once they aren't.
• In the docs. See Presets.

A picture that already carries its own labels, such as a labelled diagram, schematic, infographic or how-to sheet, is now shown whole, without our labels on top.

• Shown as it is. The picture appears in full and at full brightness, with no callouts, lines, title, caption or text plates drawn over it. It fades in, holds still and fades out, so its own labels stay readable.
• In your video's language. A picture whose labels are in another language isn't used, just like a picture with captions in another language.
• Callouts point at the right part. On other pictures, each callout now sits on the part it names. A callout whose part isn't in the picture is left out.
• A Plain Picture switch. In the editor, a Callout Diagram scene has a Plain Picture switch: on shows the picture alone, off brings its callouts back.
• Older videos: turn Plain Picture on for the scene and export again.

The VidNext team can now write to you first, and support emails only come when you need them.

• Messages from the team. News that matters to you, or a note about one of your videos, now arrives as a conversation in Help. A strip under the top bar says Message from the VidNext team, with Open. It shows on phones too. You don't have to reply, but if you do, a person reads it. See Messages from the team.
• No email for what you've already read. A support reply arrives by email only if you haven't read it in the app within a couple of minutes. Several replies sent close together come as one email.
• Your choices still apply. These emails follow the Support replies switch in Settings → Notifications. A message sent to many people at once also needs Tips & product news on.

A channel can now carry your logo, and VidNext puts it on that channel's videos.

• Branding, in the channel settings. A new Branding section takes your logo (PNG, JPG, WebP or SVG, up to 5 MB) and shows it on a video and on a Short. Set the corner and size separately for videos and Shorts, the opacity, and a fade-in at the start. Show the logo in new videos decides whether new videos get it. See Adding and editing a channel.
• Per video, on the create page. The Logo pill in the prompt box turns the channel's logo on or off for one video, and its panel changes the corner, size, opacity and fade-in for that video only. Another logo puts a different logo on one video, and without a channel Logo + takes a logo uploaded just for it. Videos that Autopilot or one-click generation start follow the channel's setting.
• Clear of everything else. The logo stays above captions, moves out of the way of bottom or top captions and keeps clear of YouTube's buttons on Shorts. A presenter window or corner text moves to another corner, and titles along the logo's edge sit just clear of it.
• In the editor. A Logo track holds the logo. Change its corner, size, opacity and fade-in, put your channel's current logo back with Use channel logo, or drag a logo from the Images tab's new Logo source — any picture dropped on the Logo track becomes a logo. See The Logo track.
• Your old videos stay as they are. Replacing or removing a logo doesn't change videos that are already made. In the thumbnail brand kit, the old Logo URL field is replaced by your uploaded logo.

A video made mostly of dark designs or night scenes now finishes normally.

• What changed. The last check before delivery could mistake a dark video for an empty one and stop it with "nothing was rendered on screen", even though every scene was there. It now stops a video only when the screen is truly empty.
• A video that stopped this way can be finished by resuming its final video step from the Resume/Redo menu. See Redo & regenerate steps.

Videos whose stock scenes were filled with found clips export from the editor again.

• What changed. When stock footage couldn't cover a scene, the footage search filled it with a found clip, but the editor pointed that scene at a file that didn't exist. The scene looked empty in the preview, and every export of the video ended with "Export Failed". The editor now finds the clip where it is saved, and the videos this affected were repaired: open the video in the editor and export it again.
• When a file can't be found, the export now stops within seconds instead of after a long render, and a panel under the Export button shows where the file is on the timeline. See Export & publishing.
• Failed exports aren't charged.

Stat cards, key-fact lists, source cards and text overlays in long videos now show only what your script says.

• No made-up numbers or names on screen. A figure, date, place, source or quote appears on a graphic only when the narration states it. VidNext no longer works out new figures from yours, such as a remainder or a currency conversion, and no longer adds sources, social-media posts or like counts your script doesn't mention.
• When your script has no figure for a data graphic, the scene shows the script's own words or stays on footage instead.
• On-screen text follows your video's language, overlay labels included.

Music video mode now carries an Experimental badge.

• Where you see it. On the Music video option in the narration source menu and at the top of the song card, which also says in one line what it means.
• What it means. Play with the mode and have fun, but results can be unpredictable — we can't promise a video will turn out the way you imagine it. See Music video mode.

Saving a preset now keeps the story images and the thumbnail setup you were using.

• Story images. The preset keeps its own copy of the pictures, with their descriptions, how each one is used, their design and their order. Load the preset and they're back. Every video made from it gets them — on the create page, from the Channel Planner's one-click Start now, and by Autopilot. Editing or removing an image on the create page changes only the video you're making; the preset keeps its images until you save the preset again.
• Thumbnails. The preset remembers whether thumbnails are on, the style, your instruction, the reference images and Use my prompt exactly. Start now makes the preset's thumbnails (its price box notes they're charged on top), and when Autopilot's Make a thumbnail is on it uses the preset's thumbnail style instead of the channel's default one.
• "Save preset?" only when something changed. The create page used to ask before almost every video, even when you hadn't touched the loaded preset, and Project settings then said the settings were changed before creating. Both now happen only when you really changed a setting.
• In the docs. See Presets, Thumbnails and Autopilot.

If you're trying VidNext with a promo code, a few short emails now help you get the most out of it. And some codes now work only through the link they were shared with.

• Right after the code. An email confirms your credits and the date your trial runs until, with three steps for a first video.
• Two days in. If you haven't started a video yet, you get three ready-made topics you can paste in.
• After your first video. Once a video is finished, a note points to the editor, the YouTube tab and Shorts.
• Before and after the end. Two days before the trial ends, and once it has ended, as before.
• They stop when you choose a plan. Nobody with a plan gets them. The credit notices follow your Credit balance setting and the two tips follow Tips & product news in Settings → Notifications.
• Codes for one link. A code shared in an ad or a partner's post may work only when you arrive through that link. If the promo field says so, open the link you got the code with, then sign up or sign in. See Redeem a promo code.

A video no longer sits on one step when an AI service stops answering.

• What changed. Every request VidNext sends to an AI service has a time limit. Sometimes a request neither answered nor failed when its time was up, and the video waited on it. Writing the script could stay on the same step for hours. Now such a request is dropped shortly after its time limit, and the step carries on the way it does after any other timeout.
• Script writing restarts itself. If writing the script shows no progress for a long time, the step now restarts automatically, the same way footage search already did.
• A video that is still stuck can be continued from the Resume/Redo menu. See Redo & regenerate steps.

The footage search in long videos gets going faster. Before it searches YouTube, VidNext groups the scenes that can share one search and double-checks each group. On some videos that planning could get stuck retrying for up to 13 minutes. It now finishes in about two minutes, so the search for your clips starts sooner.

A long video now follows a prompt that forbids stock videos, even if your content mix includes stock.

• No stock scenes are planned. The Stock Video share of your mix is split between your other types in the same proportions. For example, 40 % clips, 37 % images, 5 % motion design and 18 % stock becomes 49 % clips, 45 % images and 6 % motion design.
• Stock is never used as a fallback either. When a scene's footage can't be found, VidNext tries other video clips, then images or motion design.
• Prompts that only limit stock, such as "minimal stock footage", keep the mix you set.

Resuming a video keeps the work that already finished.

• Resume from video search keeps the footage it found. Only the scenes that still have no footage are searched, so a search that stopped near the end finishes in minutes instead of starting over. Redo still searches every scene again when you want new footage for the whole video.
• Resume from a later step starts where the video really stopped. If an earlier step never finished, VidNext completes it first. Before, a video resumed from clip download after its search had stopped could wait at that step without moving on.

On busy days, videos no longer wait in line between steps. When many long videos were started at the same time, a finished step could wait up to an hour before the next one began. VidNext now has room for many more steps at once, so the next step starts as soon as the one before it is done.

• Clip checks in long videos finish much sooner. A clip with a small issue — a large banner, side bars, or footage showing something a little different — is still swapped for a clean clip the scene already has, but the check no longer spends extra time searching the web for a new one. Someone else's video, adverts and logos your prompt bans are still searched out and replaced as before.

Every Pexels clip and photo now shows who made it, and each finished video lists its stock credits.

• In the editor. Stock video and photo results name their creator and link to their page on Pexels. What you add from them keeps that credit.
• On the project page. The YouTube tab lists the Pexels clips and photos in your video under Stock credits, with their creators. Copy copies the list for your video description.
• Add to description, when you want it. Add to description puts the credits at the end of the YouTube description; click it again to take them out. Credits are never added to a description on their own.

See Stock credits.

When a scene has no picture to show, it becomes a text card that types what your narrator is saying. The captions showed the very same words at the same time, and with centered captions they sat right on top of the card's own line, so neither could be read.

• Each word shows once. While a text card is on screen, the captions skip the words the card is already showing, whichever caption position you chose. Every other word keeps its caption.
• It follows your edits. In the editor, replace or delete the card and its captions come back by themselves, with no Resync needed. Moving the captions or pressing Resync keeps the rule too.
• Bottom captions sit where they belong. Captions set to Bottom were drawn right on the lower edge of the frame. They now sit a little above it.

Videos made before this fix change the next time they are rendered: export again from the editor, or choose scene assembly in the Resume/Redo menu.

• In the docs. See Captions & styles.

A slow AI service no longer turns a one-hour video into several hours. When one of the AI services VidNext uses answered slowly, a single stuck request could restart the whole planning step from the beginning. A 20-minute video could then take four hours or more instead of about one.

• Planning keeps its progress. Every finished part of the plan is kept. If one part fails, only that part is made again, and a retry continues where the last attempt stopped.
• Long answers are no longer cut off and thrown away. Script steps that need a long answer get the room they need. If an answer is still cut short, the next option takes over at once instead of the same request being sent again.
• Japanese narration prepares faster. The reading checks for kanji and numbers now run several at a time instead of one after another.
• Clip search starts sooner. The search for video clips starts right away for every scene that stock footage can't cover, instead of waiting for the stock search to finish. Missing clips are looked for while the rest of the video is still being prepared.

Your videos are made with the same settings, the same checks and the same pricing as before.

The length at the top of a project page is now the finished video's. It used to show the length you asked for — for example 627s — even when the finished video ran a little longer. Now a finished video shows its real length, like 10:58, the same length a price per minute is charged on. While the video is being made, you see the length you asked for with a ~ in front, like ~10:27.

• In the docs. See Pipeline overview and Credits.

We updated the Terms of Service; the previous version stays on the Terms history.

• Respect for our team. Abusive, threatening or harassing messages to our staff, our support assistant or other users break the Terms.
• Download-only accounts. Instead of a full suspension, we may limit an account to downloads: it can still sign in and download its finished videos. While an account is limited or suspended, its plan stops renewing. Restricted and suspended accounts
• What you pay for. Like the Refund Policy, the Terms now say it plainly: you pay for AI processing, not a guaranteed creative result — liking a result or not is a matter of taste, not a defect.
• The Terms box at checkout. Stripe's payment page asks you to tick that you agree to the Terms of Service and the Refund Policy, and the line under our buy buttons says the same. Pricing & plans

Switching Automatic ideas on or off from a channel's Overview no longer resets its other settings. Before, the switch put the share of trending ideas, the formats, the focus and how far ahead it plans back to their defaults. Now it changes only the switch. If you had set these before, check them once in the Idea Studio.

• In the docs. See Automatic ideas.

The video follows more of what your prompt asks for.

• The ending you describe. If your prompt says how the video should end — "end on the strongest close-up of the machine at work" — the closing shot follows it instead of a calm general view.
• Footage first when stills are off. With image search and AI images both at 0 % in the content mix, or a prompt that asks for moving footage only, a scene that loses its clip gets other footage from your video before any photo, and no AI image is generated.
• End screens are replaced with footage. A clip that turns out to be only an end screen, credits or a logo card is replaced with another clip of the scene before a photo is considered.
• Comparison labels fit. Overlap (Venn) comparisons now wrap and fit their labels instead of cutting them off, and a comparison whose values are too long for a table or card layout switches to one that shows them in full.

Adding your own pictures now works wherever you add them.

• Create page. Story images no longer fail with "Project not found". They wait with your form until you press Generate, so no empty draft appears in your project list any more. A file that is too large or isn't a JPG, PNG, GIF or WebP gets a clear message right under the upload box.
• Editor. Upload in the Images tab, Upload Your Own Image on an image scene and Upload video on a clip now work. What you upload there is also saved to your Assets library, so you can reuse it in other videos.
• In the docs. See Scripts & prompting and Timeline editing.

Japanese text in motion-design scenes now stays inside its layout. A long title next to a picture used to run as one line straight across the picture; it now wraps onto balanced lines at a size that fits its column. The same fix covers quotes, picture captions, chapter titles, document close-ups, labels around a product or a diagram, letter-by-letter titles, comparison cards, map cards and the vertical Shorts layouts — no text runs off the frame or behind a picture.

Lines now break between phrases instead of inside a word: ロシア and システム stay whole, and no line starts with ー, 、 or 。. Sentences keep the size their layout was designed for instead of shrinking to about half of it, and text is always upright: layouts that use italics for English no longer slant kanji and kana.

Typed-out narration lines no longer show a space between Japanese words (ミスと いう言葉 now reads ミスという言葉), and a long Japanese text overlay stays on screen long enough to read. Maps label countries in Japanese — ロシア, ウクライナ, フランス. Chinese and Korean videos get the same fixes, with Korean lines breaking only between words.

The fixes apply to videos rendered from now on. To update a finished video, open it in the editor and click Export Video: its scenes are drawn again with the new layout, line breaks, spacing and map names. How long a text overlay stays on screen is set when a video is created, so that part applies to new videos.

A script waiting for your review now gets one email, not one a day. It arrives a couple of hours after the script is ready if you've stepped away, and that's the only one for that script. The "don't forget your video" reminder now comes only once, for the first finished video you leave waiting — never for a video you've downloaded or sent to YouTube.

Approving a script that waited more than a day now works. Before, approving it could leave the video stuck on "processing". Now the video goes on from your approved script, and if it can't start, the script stays in review so you can approve it again.

• In the docs. See Email notifications and Reviewing before production.

The Planner now looks like the rest of your account — Team, Avatars and Pricing.

• One header on every Planner tab. Calendar, Board, Stats and Channels share the same header and tabs, on the same calm dark background as the Pricing page — switching tabs no longer flashes a loading screen. The Channels tab shows how many channels you run. Channel Planner
• The Channels tab is a library. Filter by Needs attention, Paused or Archived, search by name, handle or niche, and switch between cards and a list. Each card shows subscribers, videos and views, the runway as a bar, what's coming up and the next post; its ⋯ menu edits, pauses, archives, restores or deletes the channel. The Channels page
• Adding and editing a channel opens the same panel in the new look.
• Generate video uses your default preset when the channel has no preset of its own — on the board, the calendar and a post's panel. Channel Planner

Ask the VidNext Assistant about a discount and it now checks what is running for your account right now — a plan on promotion, the first-month price your account would pay, an upgrade offer — and quotes the same prices the pricing page shows.

• Only real offers. It no longer repeats discounts from older pages: if there is no discount for what you asked about, it says so.
• When it can't check. If the offers can't be read at that moment, your question goes straight to a person from the team instead of getting a guess.

A line written in capitals is no longer spelled out. Voices used to read some capitalised words letter by letter — "IS RUSSIA TESTING THE BALTICS BEFORE SOMETHING BIGGER?" came out as "…the B-A-L-T-I-C-S…". Lines and words in capitals are now read as normal words, while real acronyms like NATO, FBI or USA keep their usual reading.

• Captions don't change. Captions and on-screen text keep your script exactly as written, capitals included.
• Every voice and format — landscape videos, Shorts and narration regenerated in the editor.
• Existing videos keep their narration. Regenerate the narration to hear the fix.
• In the docs. How the script is read.

A promo code now takes off exactly the percent it states on a plan's first monthly payment. A 90% code gives 90% off — on the pricing page, at Stripe checkout and on the invoice. Before, plan checkout cut bigger codes to 20%.

• The first-purchase offer too. When one is running it comes off your first monthly plan payment at its full percent.

• Partner links. A link's discount comes off a customer's first monthly payment exactly as the partner set it, and the partner earns the rest of their pool. When another code or the first-purchase offer is bigger, the customer gets that one, and it counts against the partner's pool only up to 20%.

• Unchanged. Codes still never apply to yearly billing, Enterprise or top-ups; the first month is in whole dollars, rounded down in your favor; and discounts never stack — you get the biggest one.

• In the docs. See Pricing & plans and Affiliate program.

The dashboard has the pricing page's backdrop: dark ink with a soft warm glow in the top-left and bottom-right corners. The moving lights behind your project cards are gone, so the page stays still while you scroll.

The top bar is the same size on every page. On the pricing page and on the Billing, Team and Affiliate program pages it used to look slightly larger than on the dashboard and the rest of the app.

• In the docs. Getting started and the billing pages show the new look.

A redo from the video plan, the narration or the video search now downloads new clips for the new scenes. Before, a redo of a finished video from one of those steps could skip the clip download entirely, so its footage scenes fell back to photos, and a scene could pick up a clip that the previous version used for a different scene.

• What's affected. Redos from the video plan, narration or video search steps of a video that had finished before. First-time generation worked as before.
• Clip download redo now downloads the clips again instead of reusing the files from the earlier run.
• No leftovers from the earlier version in the render. For a few hours after a redo, the render could still receive the earlier version's narration or clips — a Short could lose its last line or show an old clip. Every render now uses the files of the current version.
• Existing videos are unchanged. Redo the step again to get fresh clips.
• In the docs. Redo & regenerate steps.

Footage keeps moving to the end of each line. When the clips planned for a line end before the narration does, they now keep playing from their own shots instead of stopping on a blurred still frame, and a clip can give a second, different moment of the same footage. If the footage really runs out, a full-screen photo of the subject covers the rest of the line.

• Designs get the time they need. A comparison, several-picture layout or end card is only used on a line long enough to read it — a very short line gets a picture or a text title instead — and a text card no longer holds the screen for many seconds.
• Titles fit the vertical frame. Long text overlays shrink or wrap so no letter is cut at the edge, and the big opening text is centred.
• Stock footage busy? Lines planned for stock footage look for matching video clips before settling for a photo.
• Maps. A country or region named on a map now lights up correctly, including enclaves such as Kaliningrad, and very large countries frame the part the video is about.
• Existing videos keep their footage. To swap a scene, use Find Different Clip in the editor.
• In the docs. Formats — Shorts describes how footage covers each line.

Thumb Lab now checks your credits before it starts a run. If your balance can't cover it, you see the credits the run needs and the credits you have, and nothing is generated until you top up. Before, a run could start on an empty balance.

• In the docs. See Thumbnails (Thumb Lab).

Referrals and affiliates now earn on plans. Every plan payment and top-up your referrals make in their first 12 months counts.

• Plain referrals. You earn 10% in credits of what each purchase buys — a month of their plan, a top-up, or for a yearly plan the whole year's credits at once.

• Affiliate tiers for new partners. Starter 10% and Partner 15%, with Ambassador above them. Partners and Ambassadors who reached their tier earlier keep the pool they have, and moving up a tier never lowers a pool.

• Credits pay more. Claim your earnings as credits and you get 50% on top.

• How plans pay. A plan payment's commission is counted on at most $399, the price of Pro. The first payment pays your pool minus the discount that set its price; renewals, upgrade charges and top-ups pay the whole pool.

• Yearly plans. A fixed bonus of half a month of the plan's price (Starter) or three quarters (Partner), and more for Ambassadors, paid in 12 monthly parts while the plan stays paid. Parts still to come show as Scheduled; if the customer cancels, or the payment is refunded or charged back, they stop.

• Customer discounts. A link's discount comes off a customer's first monthly plan payment only, at most 20%; links set higher give 20%.

• Enterprise plans, their top-ups and custom Enterprise deals don't earn rewards.

• In the docs. See Affiliate program; the rules are in the Terms of Service, section 11.

VidNext is now sold as monthly plans. Every plan includes minutes of finished video each month — from 30 minutes on Starter to 6,250 on Enterprise — and one minute at default settings is 50 credits. Pay yearly and save 10% (every plan except Enterprise) — whole dollars, from $44 a month on Starter: one payment, with the minutes still arriving month by month.

• Plan minutes come first. Each month's plan credits are used before any other credits on your balance, and what's left ends with the month.

• Top-ups. With an active plan, add any number of minutes from the plan card on the Credits page — from 10 minutes, in steps of 10, up to 1,000 at a time — at your plan's price per minute plus 20% (at least $1 a minute), priced per 10 minutes in whole dollars. Top-up credits never expire; you use them while your plan is active, and if it ends they wait for your next one.

• Your plan on the Credits page. A new plan card shows the plan, when it renews, the plan credits left this month, the credits that never expire, Manage billing (card, invoices, change or cancel), Renew plan when a cancelled plan hasn't ended yet, and the top-ups.

• If a payment fails, a banner in the app and an email tell you, and Update card on the plan card fixes it. Your videos stay available to watch, edit, download and export.

• First-month discount. The first monthly payment of your first plan can come with up to 20% off — a first-purchase offer or a partner's code — even if you bought credit packs before; the first month is in whole dollars, rounded down. It doesn't apply to yearly billing or Enterprise. A plan can also run its own promotion on the first monthly payment of your first plan — the pricing page marks it on the slider and beside the plan's name — and discounts never add up. Yearly billing keeps its 10% off instead.

• Nothing you have is lost. Credits you bought before never expire, accounts created before today keep creating with their balance, with or without a plan, and keep their 50 GB of storage. Accounts created from today get their storage with their plan and need an active plan to start new work — videos, thumbnails (Thumb Lab and AI edits included), avatars and Idea Studio runs; watching, editing, downloading and exporting never need one.

• Ended plans. When a plan ends, its card says so and Choose a plan starts a new one. The credits menu in the top bar links straight to Top up and Manage subscription.

• A refunded plan ends right away. If a plan's latest payment is refunded in full, the plan ends at once and the credits that payment added are taken back, unused plan credits first.

• Packs are no longer sold. The pricing page shows the plans on one slider: drag it to a plan and the price changes with it, while the card beside it shows what the plan gives — minutes a month, the price per finished minute, credits a month, videos at once, team seats, storage and top-ups; the Monthly / Yearly switch sits above it.

• In the docs. See Pricing & plans and How credits work.

A video now costs its price per minute × the length of the finished video. At default settings the price is 50 credits a minute, so a 10-minute video is 500 credits.

• The price before you start. The sticky bar on the create page shows Price: 50 credits/min × 10 min ≈ 500 credits; Details shows what each of your settings adds or takes off, and Reset price settings puts them back to their defaults.

• What changes the price. A premium voice or Enhanced Fact-Check costs a little more; your own script, your own narration, a song, a slower voice speed or your own voice API key cost less (the key's discount applies whenever your key narrates the video); your own content mix can move it either way, except in a music video, which has one price per minute whatever footage you pick; an AI avatar, thumbnails made with the video and each Short add to it. Settings VidNext picks for you never change the price.

• You pay for what you get. When you press Generate the price is set aside (up to 30% more, never more than your balance); when the video is ready you pay for its finished length and the rest comes back. A video that fails costs nothing.

• Free first redo and first export. Each video includes one free redo and one free editor export; later ones cost what they use.

• On the Credits page. Your price per minute and roughly how many minutes of video your balance makes.

• The planner's Generate shows the same price before it starts a video.

• In the docs. See How credits work.

Some promo codes now add credits to your account — and, if you don't have a plan yet, a few days to create without one. Partners and our campaigns share them.

• Where to enter one. Under Have a promo code? on your Billing page, or in the promo field at the bottom of the pricing page. A code in a link is applied for you when you sign up.

• What you get. The code's credits, at once — they never expire. If your account needs a plan to create, the code also starts a trial: for the days it gives, you can create videos, thumbnails and avatars without a plan. Your plan card and a notice at the top of the app show until when.

• When the trial ends. A plan is needed to create again. Your videos stay available to watch, edit, download and export, and your credits stay on your account — they work again as soon as your plan starts.

• The fine print. Each code works once per account, and many are for people new to VidNext: one trial per person. A code can end on a date or after a number of people have used it.

• The other codes still work the same way. The Billing page's field also takes discount codes (saved for your next checkout) and a partner's referral code.

• In the docs. See How credits work → Redeem a promo code and Pricing & plans → Trying VidNext with a promo code.

From time to time we'll offer subscribers a discount for moving up to a bigger plan — for example 30% off Business for people on Pro.

• Where you'll see it. When an offer is open to you, your plan card on the Credits page shows it with Upgrade with 30% off, and on the pricing page the offer's plan is marked on the slider with the offer price and the same button. We may also mention it in the bar at the top of the app or in a small window.

• What it covers. The rest of your current billing month and your first full month on the bigger plan, at the offer price — $489 instead of $699 at 30% off Business. Stripe shows today's exact amount before you confirm. After that the plan costs its usual price.

• Your minutes don't change. This month's extra minutes arrive right away, and every month brings the plan's full minutes.

• The fine print. Monthly plans, once per account for each plan, never together with another discount on your plan. Moving up through Manage billing is always at the usual price.

• In the docs. See Pricing & plans → Changing or cancelling your plan.

Where you allow advertising measurement, VidNext tells Meta when a sign-up or a purchase came from one of its ads, so the right ad gets the credit. Until now that report carried a hashed (SHA-256) version of your email address, an order reference with the amount paid, and technical data. It now also carries a hashed version of your first and last name and your country, which helps Meta match the report to the right ad. Meta never receives the name itself, only a hash it compares against its own users.

Nothing changes when marketing cookies are off: then nothing is sent. Switch them off any time in Cookie settings (in the site footer).

• The Privacy Policy (Sections 4 and 6) lists the new items.

News and offers now appear on the site in a slim bar at the top of the page and, now and then, in a small window — a sale with a countdown to the day it ends, a new feature, a code to copy.

• Close them any time. The ✕ hides the bar — on all your devices when you're signed in. A window closes with ✕, Esc, a click outside or Maybe later.

• Out of your way. Windows are kept rare. They don't open over the video editor, checkout, sign-in or legal pages, over the cookie choice, or while you're typing.

• Offers that apply to you. A message about a discount is shown to the people it applies to, with the price their checkout gives, and goes away when the offer ends.

• On the blog and the docs too, in the same bar.

• In your language. Ukrainian pages show the Ukrainian text.

• In the docs. See Email notifications → Messages on the site.

Scenes no longer show a look-alike of what is being said. When VidNext picks a video clip, a stock video or a web photo for a scene, it now checks what the candidate actually shows against the line being spoken. Something of the same kind but not the same thing is skipped — a football for a line about a tennis ball, another painting by the same artist, a flower garden for a line about mowing a lawn — and the scene moves on to its next source: another clip, a photo, or a generated image.

• Before/after and side-by-side templates. Each picture is matched to its own label, so the "after" picture shows the after state rather than the scene in general.
• Existing videos keep their footage. To swap a scene, use Find Different Clip in the editor.
• In the docs. Clips & footage sources explains how matching works.

Generate on the YouTube tab works again for videos on sensitive subjects. Some scripts, such as a first-aid video that shows how to help a choking baby, were declined by the AI safety filter. The button then spun for about five minutes and ended with "Network Error". Now, when the filter declines a script, VidNext writes the titles, description and tags from a short neutral summary of the script instead, which usually takes about 20 seconds.

• A clear answer every time. Generate and Regenerate titles always answer within a few minutes. If the filter declines even the summary, you see a plain message saying so, not a network error.
• No placeholder timestamps. For videos made from your own script, the description no longer lists automatic chapter names such as "Section 1, Section 2…" as timestamps.

The Cost and Time tabs on the project page have a new look. Each tab shows its name and its number — the credits, or how long the video took to make — and the one you have open is highlighted. The breakdowns underneath work as before.

• In the docs. The project page figures in Creating videos and Credits show the new look.

The Billing page has a new look — the same as the partner program pages: your balance as one big number, your plan in a card next to it, and one clear button for what you do most.

New: Your activity. Pick a period — today, yesterday, the last 7 days, this or last month, any month, your own range of days, or all time — and see:

• the credits you used, after any that came back;
• your videos: started, done, failed and in progress, and the minutes of finished video;
• what a finished minute cost you on average;
• where your credits went: videos, deleted videos, Thumb Lab, avatars and other tools;
• what came in: plan credits, top-ups and packs, and bonuses.

Your payments and the full credit history are still on the page, with the same actions — copy a payment's ID for support, open its receipt, or cancel an unfinished one.

The page you land on after paying has the new look too: it counts up the credits you just got and shows your new balance.

• In the docs. Credits has a new Your activity section, and its Billing page figures show the new look.

Your credit history now says what each line was for: Transaction History on the Billing page and the credits menu in the top bar. Lines that used to show technical wording now read plainly, in English and in Ukrainian:

• When a video starts: Credits set aside · video started. A restart reads · video restarted, and finishing a video that failed reads · finishing a video that failed.
• When a video is ready: Video finished · unused set-aside credits returned if it cost less than was set aside, or Video finished · it cost more than was set aside for the extra. A redo reads the same, starting with Redo finished.
• When nothing is made: Set-aside credits returned, with the reason: the video failed, was cancelled or was deleted.
• When you cancel at script review: Video cancelled at script review, showing the work done so far and the unused credits returned.

The amounts and your running balance are unchanged.

• In the docs. The Transaction History figure in Credits shows the new lines.

A video that waited at script review for more than a day now sets its credits aside again when you approve the script. After a day, the credits set aside for a waiting video go back to your balance, so they aren't locked while you decide. Approving now sets them aside again, the same way pressing Generate does, so the video always runs on credits set aside for it. Your credit history shows the line Credits set aside · script approved.

• Not enough credits when you approve? The review tells you how many credits you need. Top up, then approve again. The video keeps waiting, and nothing is charged.

• Already running as many videos as you can? Wait for one to finish, then approve.

• In the docs. Reviewing before production and How credits work.

Enterprise plans and custom Enterprise deals don't earn referral or affiliate rewards. Nothing changes for the purchases you earn on today: every pack a referral buys within their earning window still counts.

• On the affiliate page. The FAQ answer to What do I earn on? says so.
• In the docs. See How earnings work.
• In the Terms. Section 11.2 (Rewards) says so; the Terms are dated September 29, 2026.

The Get More Storage card is gone from the Storage page (account menu → Storage). Its Subscribe button could not complete a purchase, so the card is removed rather than left looking like it works. Everything else on the page stays the same: your usage, the list of projects by size, the quick selections, and Clean up / Delete.

• In the docs. Storage no longer describes the card.

A video could stop at the Generate Infographics step when the AI wrote an animated scene in the wrong format. Most often it was a numbered list whose lines came back as plain text instead of list items. The step now always finishes:

• Scenes keep their design. An answer that doesn't fit the scene's design is redone before it's used, so the numbered list or feature list shows up as planned.
• One scene can't stop the video. If a scene still can't be built, it's shown as a text card and the rest of the video carries on.

Illustrated presenters. A presenter can now be a cartoon, anime or 3D character. When you generate a persona, choose Illustrated and an art style. To upload one, turn on Illustrated character in the Upload tab. The character keeps its art style in every version, shot and clip. Existing characters from films, series, games or brands, and caricatures of real people, are declined.

The avatar's voice. The Voice tab of an avatar is now the same voice picker as the create page, with the providers available to your account and each voice's settings (Cartesia speed, Gemini style and pace, Fish Audio tags). Picking the avatar on the create page applies the voice and its settings, and preview clips speak with it. When the avatar has its own voice and the video narrates with another, the AI Avatar card says so and offers to use the avatar's voice.

Fixes

• Candidate portraits show a tile per portrait with its status from the start, also after a reload. Picking one before all four are done no longer leaves the avatar stuck as a draft. Portraits not started yet are skipped and not charged, and one that was already rendering can be added as a version.
• Preview clips on several versions each keep their own progress, also after a reload. A second press no longer fails with "speech synthesis produced no audio file".
• A photo with a transparent background is placed on a plain studio backdrop. The animation no longer shows an outline around the person.
• You can drag a photo onto the upload picture.
• In the editor, dragging a presenter from the Avatars tab onto a scene works again, and a drop that cannot add one says why.

Read more in AI avatar presenter and Voices.

Avatar videos. Write a script — or let AI write it in your presenter's character — pick an avatar, one or more of its looks and a voice, and get a talking video with the voice in it. Every Generate keeps the earlier takes, and each look you pick becomes its own take, so you can compare. Use the avatar's own voice, pick another one for that video, or upload your own recording for the presenter to lip-sync to. Download a take with voice, without sound, or the voice alone. A take is charged once when it is ready; a take that fails or that you stop is free. Avatar videos are on the new Videos tab of the Avatars page when they are available for your account.

Drag them into any project. In the editor, the Avatars tab now has Videos and Looks. Drop an avatar video on the Video lane for a clip with its voice, or on the Avatar track for a presenter window that speaks — nothing is generated. Its voice switch and volume are in the element's properties.

A new Avatars page. Your avatars are a grid of cards, each showing its main look and two more; hover a card to play its preview. Search, filter by realistic or illustrated, and keep favorites with the heart — they come first on the page and when you pick a presenter for an avatar video. An avatar opens full width with its Looks, its Videos and Settings. Versions are now called looks, and Generate a look and Compose a set show what the new looks cost before you start. The voice has its own window, opened from the voice button at the top of the avatar's page.

Read more in AI avatar presenter.

Narration with Cartesia voices, including the SONA PRO collection, is now generated with the newest version of the voice model. Every part of a video is voiced with the same version, also when a part has to be voiced a second time.

Finished videos keep their narration. To voice a part again with the new model, open the video in the editor, use Regenerate Narration on its narration and re-render.

See every shortcut. Click the keyboard icon at the top right of the editor, or press ?, to open the list of every keyboard shortcut and mouse action: playback, editing, selection and timeline zoom. On a Mac it shows ⌘, elsewhere Ctrl, and the list stays open while you try the keys.

Undo you can rely on. Ctrl+Z (⌘Z on a Mac) now waits for your last change to save before it undoes, so pressing it right after a move or a trim undoes that change rather than an earlier one, and quick presses undo one step each, in order. Undo, redo and the letter shortcuts now also work right after you drag a slider, with Caps Lock on, and with a Ukrainian keyboard layout, and Ctrl+Shift+Z redoes on Windows too.

Smaller fixes. Ctrl+D (⌘D) duplicates the selected clip, as the right-click menu says. Keys no longer change the timeline behind an open dialog, and Esc there no longer clears your selection. Holding a key no longer repeats a split, a delete or play/pause. Ctrl+− zooms out after "fit to view" on a long video too. Esc leaves a text field, so the next key is a shortcut again, and after you click the YouTube preview the shortcuts work again as soon as the mouse is back over the editor.

Read more in Editor tour.

Japanese narration now says words whose kanji can be read more than one way with the reading the sentence needs: 弾薬庫は空になり is voiced からになり ("empty"), not そらになり ("sky"), and 損害を被る is こうむる, not かぶる. Rare words, technical terms and names written in kanji get their reading too, so the voice no longer has to guess them.

Before each part of the narration is voiced, the script is checked for the words the voice could misread, and those words are voiced from their reading. Captions and on-screen text still show your script exactly as written.

The fix applies to narration generated from now on. To update a finished video, open it in the editor, use Regenerate Narration on its narration and re-render.

The top bar has the new VidNext logo, and it takes less room: the links sit a little closer together and the language button is now just the flag. Everything still fits on one line on smaller laptop screens, and your profile picture no longer gets squeezed when the bar is full.

Avatars carries a small New badge for a few weeks — it's where you create AI presenters and avatar videos.

See the top bar in How credits work.

Team → Projects now shows which channel each video was made for.

• Channels. The table has a Channel column, a Channel filter (each person's channels under their name) and a sortable Channel heading. Click a channel's name to open it read-only. You see its numbers, top videos, AI insights and tracked competitors on Overview, and all its YouTube videos with their stats on Videos. Nothing there can be changed or charged.
• Search finds more. Search on the Projects tab and on your dashboard now matches any word of a video's title, its other title ideas, its prompt, its YouTube description, the planner idea it came from, or its channel's name. Before, the dashboard only searched prompts.
• Credits on the day the video started. A video's cost now counts on the day it was started, including the unused credits that come back when it finishes. Before, a video started late in the evening that finished after midnight had its refund counted on the next day, so that day could show refunded instead of what the videos cost. Other tools still count on the day you use them.
• Members are told. The Team page, the invitation email and the page where you accept an invitation now also say your team owner and team leads can see your planner channels.

Read more in Teams.

If VidNext sets up a team for your company, your members are already on the Members tab, with no invitation needed. They start on Own credits, so nothing changes in who pays until you switch someone to Team credits. You can then set their daily, weekly and monthly limits too. Members see the team, and who can see their videos, on their own Team page.

Read more in Teams.

The top tier of the Affiliate program is now by invitation. Starter and Partner work exactly as before.

• On the affiliate page. The tier cards show Ambassador as Invite only: its pool and terms are shared with the partners we invite, and the card has an address to write to if you think it's you. In the earnings calculator Ambassador has a padlock and can't be picked. Click it to see why.
• In your hub. The Program tab marks Ambassador By invitation with a padlock, and the tier strip on the Overview tab says so, instead of showing bars toward it. If you are already an Ambassador, nothing changes: your pool, your links and your splits stay as they are.
• Partner's bar is unchanged: $150 in net revenue or 3 new paying customers in your last 90 days. See Tiers and splits.

When a video's generation fails, the credits held for it come back in full. If you then finish that video with Redo or Continue, the finished video is now charged once: its actual cost, never more than the amount held when you created it. Continuing holds that amount again, so it needs to be available on your balance. Previously such a redo was billed only for the steps it re-ran.

The project page has a new Settings button in its top bar. It opens Project settings: the preset, the prompt or script, the voice, the writing style, the visuals, the music and ending, the AI avatar and the thumbnails the video was created with. The button is there while a video is still being made, too.

• The top of the window sums the video up: its status (and what went wrong, if it failed), when it was created, its project ID with a copy button, chips for the preset, mode, format, language, voice and models, and tiles for the credits charged, the length against its target, the number of scenes, the script's word count and how long the video took to make.
• Two tabs: Settings lists every setting in groups, and Prompt & script shows the prompt, the script and your other texts, each with its own copy button. Long scripts are folded behind Show full script.
• Copy all copies the summary and every setting as text, which helps when you want to repeat a video's setup or show someone how it was made.
• Settings VidNext chose for you are marked System default or Picked automatically, and settings left on Auto show what was picked once the video was planned.
• The preset label says when the preset was changed before creating, when it came from your team, or when it has since been deleted.
• Older videos show what could be recovered, and their preset shows as Preset not recorded.
• Team owners open the same window for their members' videos from Team → Projects.

Also: on the Ukrainian create page, the footage source names are now translated.

Read more in Pipeline overview.

The Team page has a new look, in the style of the partner hub, and now has three tabs: Members, Projects and Sharing.

• Projects. Team owners see each member's videos with their status, length and what each one was charged, with totals per person and for the whole team for today, yesterday, the last 7 days, this month, last month, all time, a month you pick or your own dates. Filter by person or status, or search. Open shows a video read-only: you can watch it, download it, and see what it cost and the settings it was made with, but you can't edit, redo or generate anything for it.
• Deleted videos. The Projects tab also counts videos that were deleted: how many each person has in the period and what they were charged, on the people cards and in the totals. Choose Deleted under Status, or Show in the note above the table, to list them with when and by whom each one was deleted.
• Sharing. Team owners can share Presets (video presets, thumbnail styles and writing styles), Voices, API keys, and Music & avatars with their members. Nothing is shared until you turn it on. Members find shared items in Team groups wherever they pick them, and can copy presets and thumbnail styles to their own. Members never see your keys, and narration made with your key gets the same 50% own-key discount. Turning a switch off stops members picking that item for new videos; turning off API keys stops your keys being used right away.
• Who can see your videos. If you're on a team, the Team page now says who can see your videos and what they cost: your team owner, and any team lead VidNext set up for your company. The invitation email and the page where you accept it say so too.
• Credits and limits per member. On the Members tab, a team owner chooses for each member whether their videos use the owner's credits (the default) or their own, and can cap what a member spends of the owner's credits per day, per week and per month, with what they've spent so far. At a limit, new videos and tools wait until it resets — a video that's already being made always finishes. Members see whose credits they use and what's left of their limits.
• Fixes. Leave team opens its confirmation again, so leaving a team works; before, pressing it did nothing. Dates on the Team page now follow your language.

Read more in Teams.

When you edit a thumbnail variant, you can now attach up to 9 reference images — a logo, a face, a product shot — instead of 2. The edit dialog shows how many you can add.

In some browsers the top bar could blink, most of all while you scrolled the dashboard. Its see-through, blurred background was redrawn constantly over the page's drifting light and sometimes flashed. A few pop-up windows and panels had the same problem.

• A solid top bar. The top bar is now solid on every screen size, so it stays steady while you scroll. Phones already had the solid bar.
• The project header too. The header at the top of a project page, with the status, Redo and the editor, got the same fix.
• Pop-up windows. The YouTube Guide, the cookie notice and settings, and the editor's voice picker and Stop this render? window now dim the page behind them instead of blurring it. The YouTube Guide now also covers the top bar, like every other window.
• Other bars and panels. The selection bar on the Storage page, the pricing page's pack slider and promo field, and the pipeline panel on a project page no longer blur what is behind them.

A custom script can run up to 60 minutes of narration — about 54,000 characters in most languages, or about 23,000 in Japanese, Chinese or Korean. The script panel shows the estimated length as you type. When a script is over the limit, Generate stays off and the bar at the bottom says how long the script runs, so you can shorten it or split it into two videos. The same limit applies when you edit a script in Review script before approving it.

The length and credit estimates for Japanese, Chinese and Korean scripts are now accurate. They used to count those characters like letters and showed less than half the real length.

Fish Audio voices now speak with Fish's newest and most expressive voice model. If it's ever unavailable, narration switches to the previous model on its own, so a video never stops because of it.

Fish Audio voices you cloned yourself now narrate videos even if you never added your own Fish Audio API key. Before, generation could stop at the narration step asking for a key. Regenerate Narration in the editor also works with voices cloned on your own Fish Audio account.

Captions no longer show the performance cues that direct a Fish Audio voice, such as warm and sincere. They could slip into the captions of narration that contained numbers, where the numbers themselves sometimes appeared as spelled-out words instead of digits; both are fixed.

The voice-and-music preview on the create page now plays Fish Audio voices too. Scripts for Fish Audio voices are also sized to how fast each voice actually speaks, so finished videos land closer to the length you picked.

Thumbnail variants are charged when each one is made, and the price is shown next to the generate button. When a project's thumbnails finished before the video itself, the final video charge could count those thumbnails a second time. From today the final charge leaves out thumbnails that were already paid for.

Since September 24, several AI steps counted the model's reasoning twice when working out their cost: planning, clip selection, fact checking, script revision and stock footage matching (the "stock video prescreen" line in a project's Cost tab). They now count it once, so these steps cost less from today on.

Stock footage matching also does less repeated work. When a scene needs a second search, only the new clips are checked; clips already checked for that scene are no longer checked again. Generic b-roll scenes that try stock footage before other sources are now checked together instead of one by one, so a clip that fits several of them is looked at once. And the same stock clip is no longer used twice in one video, including between stock scenes and those generic b-roll scenes.

The Cost tab also lists every AI call of a project now: calls that finished at the same moment could be left out of it before.

Japanese narration no longer reads years digit by digit: 2026 is now said にせんにじゅうろく, and numbered units such as 第20軍 are read as whole numbers. The word next to a number — a counter, unit or prefix such as 年, 人, 万, 億ドル or 第 — is no longer said twice.

Korean narration now uses Sino-Korean numbers with dates and ordinals (8월 is read 팔월 and 5세대 오 세대) and no longer repeats a counter after a number.

Every number is now also checked by listening once the narration has been voiced. When the voice gets one wrong, that part of the narration is voiced again with the number written the way it is pronounced. If a number may still sound wrong, the editor shows it on the narration so you can listen and regenerate it.

Number cards and animated stats show a year without a thousands separator — 1997年, 2010年代, 2026년 instead of 1,997年 — while quantities such as 12,345人 keep theirs.

Captions and on-screen text still show the digits exactly as they are in your script. The fix applies to narration generated from now on; to update a finished video, open it in the editor, use Regenerate Narration on its narration and re-render.

In the channel planner, Delete idea and project now works exactly like deleting the project from your dashboard or the Storage page. All of the project's files are removed, which frees their space in your storage. A generation still in progress is stopped, and the credits held for it are returned to your balance. Before, only part of the files was removed and credits held for a running generation stayed held. Credits already spent on the project are not refunded, as the dialog says.

On a video with an AI presenter, the list of avatar scenes now lives in an Avatar tab in the right sidebar of the project page, after Thumbnails. The player, its download buttons and the Cost and Time tabs always stay in view; before, a long avatar list could push them out of sight. The tab shows how many avatar scenes the video has, with a spinner while a clip is generating and an amber dot when some scenes show the still image. Every Regenerate button shows its credit cost, and Regenerate all shows the total.

A scene whose first attempt stopped on a network error no longer shows generating forever, so Regenerate all works again. The Time tab also lists the avatar step in its place in the pipeline, after the overlays, instead of at the end.

Read more in AI avatar presenter.

Long videos no longer stop on the last spoken word. After the narration, a calm closing shot of your topic plays while the music keeps going, then the picture and the music fade out together. Choose the length in the music step — Off, 3 s, 5 s (the default) or 10 s; without music the closing shot is a short 2-second fade. Presets remember your choice.

In the editor the fade is an Ending block on the Effects track that you can resize, adjust or remove, and a video without one can get it from the new Ending button above the timeline. Shorts and music videos keep their own endings.

Read more in Background music and Timeline editing.

The AI avatar estimate on the create page is a range, and its top is now a true worst case: besides seconds × rate it includes the minimum charge every short take carries (an intro, an outro, a corner comment) and the silence padding a split take may need. The cost chip also says how many seconds a take bills at least. Once your plan is written, the credit hold is adjusted to the actual presenter clips, so the final charge stays within what you saw.

Read more in AI avatar presenter.

On presenter models that have an energy setting, the Motion choice (Still, Calm, Natural, Lively) now sets it for every take. Before, those models always ran at their default level whatever you chose. Their automatic motion prompt now describes movement only, so Still and Calm takes come out steadier.

Read more in AI avatar presenter.

Uploading a photo for an AI avatar now tells you right away whether it will work. Pick a file in the New avatar window (or in Upload a version) and the card under it shows the verdict: what the check sees, any heads-up, or — when the photo is declined — the reasons and what to change (one adult face looking at the camera, eyes visible, nothing covering the face, sharp and evenly lit, a real photo without filters, stickers, watermarks or text). Nothing is uploaded until a photo passes, and a declined photo no longer leaves a half-made avatar or version behind.

• The minimum size is now 512 px on the short side (PNG, JPEG or WEBP, up to 15 MB).
• The New avatar window and the version sheets have a new look.

Read more in AI avatar presenter.

A finished video with an AI avatar now lists every scene where the presenter appears right on its project page, under the video and the downloads: the status of each scene, a Regenerate button per scene and one for all of them. Before, the list sat on a step of the manual project page that generated videos never opened.

Read more in AI avatar presenter.

In the editor, a new take of an avatar scene used to reach the timeline only while that element's properties panel stayed open; if you had moved on, the preview kept the old clip until the page was reloaded. The editor now watches every pending take and swaps the new clip in the moment it is ready. An export started while a take is still generating keeps the previous clip instead of dropping the presenter from that scene.

Read more in AI avatar presenter.

In the editor, the Next take section of an avatar element now shows the presenter's versions as thumbnails instead of a dropdown. The still that plays now is marked Current; tap another one to use it for the next take, then press Regenerate. Versions of other avatars sit behind a fold.

Read more in AI avatar presenter.

Every "are you sure?" step now opens VidNext's own dialog instead of the browser's plain pop-up: regenerating an avatar take, deleting an avatar, a version, a narration, a clip, a thumbnail or a project, resetting the editor timeline, force-cancelling an export, removing a team member, cancelling a subscription and the rest. Escape or Cancel closes it, and destructive actions are marked in red.

In the editor, every element that plays a file — a presenter take, a video clip, an image, narration, music or a sound effect — can now be saved to your computer on its own. Right-click the clip on the timeline and choose Download file, or select it and press the download button in the properties panel header. A presenter take saves as an MP4.

Read more in Timeline.

In the editor, picking a version under Next take that was added to the avatar after the video was made, or that belongs to another avatar, silently fell back to the version the plan had used — and charged for that take. The picked version is now used, and a version that is no longer available is refused before anything is charged.

Read more in AI avatar presenter.

When you generated thumbnails for a project, Thumb Lab's project view showed the video's prompt at the top instead of the brief you wrote for the thumbnails. It now shows Your brief — the instruction you last wrote in Thumb Lab, on the project's Thumbnails tab or on the create page — and shows the video's prompt only when you haven't written one. This applies to thumbnails you generate from now on; a project whose instruction came from the create page shows it right away.

Also fixed: with a project attached, the Thumb Lab composer ignored the "Use my prompt exactly" switch and the extra-web-references switch. Both now apply, just as they do for standalone thumbnails.

Read more in Thumbnails.

Sign-in verification codes and password-reset codes now arrive from VidNext <notify@vidnext.ai>, the same sender and the same design as every other VidNext email, instead of a separate notifications@ address. Mail filters that already know the VidNext sender treat the codes the same way. If a code once landed in your spam folder, adding notify@vidnext.ai to your contacts helps later ones reach the inbox.

When an AI presenter sat in a corner window over footage and a scene transition followed, the window stayed on screen through the transition, frozen mid-word, while the scene slid or wiped away. The window now fades out just before the cut and fades back in once the transition has ended, in rendered videos and in the editor. A full-screen or split presenter still carries through the transition as before, and a cut inside one continuous presenter take still has no transition at all.

The AI avatar presenter can now appear in extra shots — not only as the portraits saved in the library. When the plan writer decides a moment needs the presenter somewhere else or doing something (walking through the server hall, at a café table with a laptop), it describes the shot, a still of it is made from the chosen version and that still is animated. It is always the same person and wardrobe; other people, brands and readable text are ruled out.

• On by an admin, capped per video. The feature is switched on per workspace; each video may ask for a limited number of distinct shots, and the create-page credit range includes them.

• Editable in the editor. Avatar scenes gain a Shot box: change the text for a new still, or clear it to return to the plain version. Save shot to library turns a shot you like into a regular version of the avatar.

• Never a broken scene. A shot that cannot be made falls back to the chosen version, and rerunning the avatar step reuses shots already made.

• Faster persona setup. The New avatar window takes a one-word topic and Autofill with AI writes the name, the brief and picks styles in your language. Style tags became Style chips: a realism level (Maximum realism, Polished, Cinematic) and a look (Studio actor, Model, Documentary host and more) that steer every portrait of the presenter.

• Wider framings. Versions can now be knee-up, the presenter at a desk or table with the surroundings in view, or full body standing. The plan writer uses them for openers and moments where the place matters, never for a corner box.

• Sets inside an avatar. Name a set — Car, Café, Office — when you generate, compose or upload versions, and the create page lets you pick the set for a video, so the presenter stays in one place while the angle changes. Without a choice the plan writer picks one set per video. Presets remember it.

• Edit a version. The pencil on a version tile opens "Edit this version": describe the change — background, setting, wardrobe, pose — and a new still is made from that version with everything else kept. Tick "make it the default" and later versions start from the edited one.

• Compose a set. On the Avatars page, make several versions at once: tick the framings and angles you want, add a setting or pose, and every combination is generated from the same base. Portrait prompts now describe the subject, the background and the light and camera in concrete terms, which makes realistic presenters noticeably more real.

• Versions that keep their word. A generated version — say a close-up, front — is now checked against the framing and angle you asked for and redone once when the picture kept the reference's crop. You pay for the delivered image only.

• Slow zoom on full-screen scenes. A full-screen presenter can now push in or pull out slowly over the scene, so a talking head does not sit like a still frame. The admin sets the default and the strength; the editor's Slow zoom buttons change it per scene.

• Motion control. Every avatar clip is now written for a motion style — Still, Calm (the default), Natural or Lively — so the presenter no longer nods, tilts and gestures on every line. The admin sets the default; the editor's Motion buttons change it for the next take.

Also new: the Light presence level between Intro & outro and Balanced — the opener, the closer and one or two key moments.

Read more in AI avatar presenter.

Two fixes for videos with an AI presenter:

• When the plan put a picture-in-picture presenter on one of the presenter's own scenes, that moment was rendered as a full-screen take instead. It now plays as intended: the scene shows the footage the plan described, with the presenter in a small window over it.
• A scene transition next to a full-screen presenter take with a slow zoom briefly showed the take without its zoom, so the picture jumped for the length of the transition. The same happened to a picture-in-picture window, which disappeared during transitions. Transitions now carry the presenter exactly as the scene shows it — same zoom, same window — on both sides of the cut, in rendered videos and in the editor.

When a presenter clip could not be generated, the finished video showed the presenter's still image for that scene, but the editor showed a black gap there. The scene now keeps its element on the Avatar track, showing the same still, so the timeline matches the video — for new videos and for existing ones the next time they are opened.

Select that element and press Generate the presenter clip to make the missing clip right there: it replaces the still in the same element when ready.

A clip that failed on a passing network error is also retried once before the scene falls back to the still image. As before, a scene on the still is never charged, and Redo avatars on the project page regenerates it.

In the channel planner's Competitors panel, a rival channel counts as Beatable from a beatability score of 45 instead of 60. Catchable now covers 30–44, and Uphill stays below 30. The Your position line and the role filter use the same bands, so more of your peers show up as beatable now rather than catchable later.

Videos can now be hosted by an AI avatar — a consistent presenter who talks to camera in some scenes, lip-synced to the narration. It is rolling out gradually: the controls appear once the feature is enabled for your account.

• Pick a presenter and a presence level on the create page. Choose a ready-made avatar or one of your own, then Intro & outro, Light, Balanced or Presenter; the plan writer decides which scenes go to camera and whether the presenter is full screen, in a corner or sharing the frame. Shorts use split screen and a circular corner box above the captions.
• Your own library. The new Avatars page generates a persona from a brief (pick one of four portraits), or takes a photo of you with a likeness attestation and an automatic check. Add versions — angles, framings, settings — bind a narration voice, and make a five-second preview.
• Its own track in the editor. Avatar scenes sit on an Avatar track with layout, corner, size and shape controls and a Regenerate button; drag a version from the library's Avatars tab onto any scene to add a presenter.
• Charged per second of clip, never for failures. The create page shows the credit range; a clip that cannot be generated shows the presenter's still image instead and can be redone from the project page (Redo avatars) at no extra charge for the failed attempt.

Read more in AI avatar presenter.

Analyze with AI in the channel editor sometimes ended with "The analysis did not come back in a usable form" although the channel had been read and a full profile written. It hit channels whose videos are not in English most — a Japanese channel, for example — and running the analysis again did not help.

• A stray label no longer sinks the whole analysis. Formats, tones, age groups, monetization and countries are chosen from fixed lists. When the AI wrote one of them in its own words — a translated or differently spelled format name — the entire result was discarded. The closest matching option is now used, and a label that matches nothing is left out.
• The backup analysis now actually runs. When the first attempt is unusable, a second one takes over. Until now that second attempt could fail on its own before answering; it now completes.
• Nothing to redo, nothing lost. Failed attempts were never charged, and they no longer count toward the daily analysis limit. Open the channel and run Analyze with AI again.

A text overlay — a name, a date, a statistic — is synced to the moment the narrator says it. When that moment came in the last second of a scene, the overlay was cut off by the scene change and showed for about a second: too short to read, so it only blinked.

• At least 2.5 seconds on screen. Every text overlay now stays up for at least 2.5 seconds. When the narrator reaches the fact late in a scene, the overlay opens a little earlier so it still gets its time and ends before the scene changes.
• Short scenes carry no overlay. A scene shorter than 4 seconds cannot hold an overlay long enough to read, so the plan's overlays for such scenes are left out — the narration still says the fact.
• Editor and video agree. The Overlays track in the editor is built with the same rule, so what you see on the timeline is what renders.
• Shorts are unchanged. Vertical videos keep their own faster overlay timing.
• Existing videos. Redo a finished video from the overlays step in the Resume/Redo menu; the assembly and final render follow automatically. Or simply delete the short overlay on the editor's Overlays track.

A "5 reasons" or "top 10" video opens with a scene that previews every item at once, numbered. In some of these videos that opener rendered as a small rank badge in a corner of an otherwise black frame for several seconds, because the planner had been handed a corner overlay design as if it were a full-screen list.

• Full-screen list or picture wall. The opener now uses a full-screen numbered list of the items, or a wall of their pictures when every item has one. Both come from the designs you selected. When your selection has neither, the video opens on real footage of the subject with the numbered overview on top.
• Overlay designs stay on top of footage. Rank badges, stat callouts and quote cards are overlays; they now never stand alone as a scene in a new plan.
• Existing videos. If one of your videos opens on a badge over a black frame, redo it from scene assembly: a picture is found for that scene and the badge is placed on top of it. "Assembly only" does not change it.

The Marketing cookie category now covers one more thing. Where you allow it, VidNext may share a hashed (SHA-256) version of your email address with its advertising providers — Meta, Google, X and Reddit — to build advertising audiences: for example so people who already have an account stop seeing VidNext ads, or to reach people similar to existing customers. The providers never receive the address itself, only a hash they match against their own users.

Because the meaning of the choice changed, everyone is asked once more. The cookie wall (European Economic Area, United Kingdom, Switzerland) or the short notice (everywhere else) appears again on your next visit, and only a choice made from today on counts for these audiences.

• To opt out, switch marketing cookies off in Cookie settings (in the site footer), or email support@vidnext.ai. Your address is then left out at the next audience update.
• The Privacy Policy (Sections 4, 5, 6 and 9) and the Cookie Policy describe the change, and the Cookie settings dialog says what the Marketing category covers.

A Short made on September 17 showed a "VS" comparison with two black halves for eight seconds: the graphic was there, its two pictures were not. Here is what happened and what changed.

• Two-picture graphics get AI pictures when your video asks for them. When your content mix favours Generated Images and a scene compares two subjects, both pictures are now generated — each written for its own subject and in your video's look. Before, only single-picture graphics were generated; the two pictures of a comparison were searched as stock photos instead, which rarely exist for a specific subject.
• A picture search that finds nothing no longer ends there. If a photo can't be found, the same search runs on the web, and if that fails too an AI picture is generated. Real people, brands and logos are the exception: they are never invented.
• Before / after graphics search for the right thing. Their two pictures are now searched by what each one should show. Before, they could be searched by the words "Before" and "After".
• A graphic that still can't get its pictures never renders empty. Graphics that are nothing without their pictures — comparisons, before / after, photo walls, picture stacks, device and product showcases — get one more search and one more AI attempt during scene assembly. If pictures still can't be produced, the scene becomes a text scene that shows what your narrator is saying.
• Last-resort text scenes are livelier. When the line is short (up to 10 words), the words appear one by one in step with the voice, in your style pack's look. Longer lines keep the typewriter card.

If one of your videos shows a graphic with empty or black panels, open it and choose scene assembly in the Resume/Redo menu: the graphic gets its pictures, and your narration and clips stay as they are.

The top of the dashboard has been tidied up, especially on phones.

• One status filter instead of two. The counters under Welcome back — Total, Completed, In Progress, Draft, Failed, Cancelled — were always the filter for your video list. The toolbar above the grid repeated the same filter as a row of icons; that row is gone.
• Nothing runs off the screen on a phone. Start Creating now takes a full row of its own, with Watch demo, the YouTube guide, Docs and Help underneath it. The counters sit in a single row you can swipe sideways, so the header stays short.
• Search gets the room. On a phone the project search stretches to fill its row; the channel picker and the per-page choice wrap beneath it.

Nothing changed in what the filters do or in the order of your videos.

Some recent videos — most of them made on September 17 — showed a full-screen text card where footage belonged, and the card typed an internal planning note about the picture instead of anything meant for viewers. It hit landscape videos and Shorts alike, including a Short's opening scene. Here is what happened and what changed.

• A scene that found its clips shows them. When a scene is built from two or more clips and they end a fraction of a second before the narration does, the scene used to be replaced by a text card — all of its footage was dropped to hide a gap shorter than a blink. The clips now always stay.
• Small gaps are filled with more of the same shot. The last clip of a scene simply plays a little longer from its own source, and it never runs into the next shot, so there is no flash of another picture. When the source has nothing more to give, the clip's last moment eases forward and back to cover the gap.
• A text card only ever types words meant for your viewers. A card appears only when a scene truly has no picture to show — normally after a web image search and an AI-generated image have both failed. It types what your narrator is saying at that moment, so it always matches your video's language. It can no longer show planning notes, image descriptions or search phrases.
• The short typed lines over a held frame follow the same rule. They come from your narration or from on-screen text planned for the scene.

If one of your videos shows a text card where a clip should be, open it and choose scene assembly in the Resume/Redo menu: your clips and narration stay as they are and the scene is rebuilt with its footage.

When you redo the narrations step, your titles, typewriter cards, timelines and other motion-design scenes now follow the scenes' new lengths.

• What went wrong before. A redo could give a scene a new length while its motion-design animation kept the old one: the animation rushed through in a fraction of a second and the rest of the scene showed an empty dark background. The editor preview looked right; only the rendered video was affected.
• What changed. At assembly, every motion-design scene whose length changed is fitted to its new length again — the same fitting the editor already does — before the video renders.
• Nothing to do on your side. Videos redone from now on come out right. If you already have such a video, redo scene assembly from the Resume/Redo menu to re-render it (the fast Assembly Only option reuses the old fitting).

The Script model block at the bottom of the Video style section — the choice between Research-grounded and Creative writing — no longer appears on the create page.

• Every new video is written by the research-grounded writer. It was already the default; the choice added a step without changing the result for most videos.
• Nothing to set up. The rest of Video style — content type, writing style, Enhanced Fact-Check and CTA placements — works exactly as before.
• Saved presets keep working. A preset saved earlier with Creative writing still applies that setting when you load it; the same goes for a project made from such a preset.

When you upload your own narration in Japanese, Chinese or Korean, the scene changes now land on the pauses in your recording — the same as they already did for other languages.

• What went wrong before. A video with such a voiceover could come out with one normal scene, a burst of very short scenes, and then a single scene held on screen until the end of the audio. It affected Shorts and landscape videos alike; text-to-speech narration was not affected.
• What changed. VidNext now reads the transcript of your recording exactly as it was spoken, phrase by phrase, so every scene gets the stretch of audio that belongs to it.
• Already have such a video? Open it and choose narrations in the Resume/Redo menu. Because the scene lengths change, the footage is cut again to match, and the video is re-rendered.

Each link & code card in the affiliate hub now keeps three amounts apart, so a big number is never mistaken for money that is already yours.

• Customers spent. What the customers a link brought have paid — every purchase, net of refunds. It replaces the old "Revenue" figure, which read like your earnings and was not.
• Earned in total. Your own share of that spend, all time, with what has already reached you underneath: paid out, credits on your balance.
• Not paid yet. The part still on its way, with a line that says where every piece stands: on hold (and the day the first amount unlocks), ready to claim, requested as cash, or credits pending. Dollars and credits sit side by side when you have both.
• Credits count per link now. Earnings that arrive as credits — every referrer's, and every affiliate's before cash payouts open — used to leave the link cards blank. They show on the card they came from.
• A 0% link says so. A link that hands the whole pool to the customer on every purchase says it earns you nothing instead of showing a number.
• Overview. Under your earnings headline, the first line now says how much is not paid yet, and the "Earned" column of your referrals includes cash you have already requested.

Details in Affiliate program.

In finished videos, the text inside some designed scenes could tremble: the rows of a comparison table, a column of numbered feature cards, a document close-up, a person's name card or a big title shifted by about a pixel while the scene was meant to be holding still. It was easiest to notice on small text, and stronger when a Film, Dusty, VHS or Analog finish was on.

• Text stays put. Once a table, list, card, chart, quote or title has arrived on screen, it no longer drifts or slowly zooms. It sits still until the scene ends, so every letter stays sharp.
• Pictures still move. Photos and backgrounds keep their slow zoom and pan. In title scenes with a cut-out subject, the subject now moves gently against a steady headline.
• Floating objects still float, smoothly. Device mockups, photo prints, pinned cards and product cut-outs keep their gentle float, without the small jumps.
• Film finishes keep their look. Grain, dust, scratches, flicker and vignette are unchanged. The finish no longer nudges the whole picture underneath it.

New videos and new exports get this automatically. A video you have already made keeps its current file; to refresh it, export it again from the editor.

A Short about one named person could end up half-filled with pictures of something else: other people's tweets, another company's comparison table, a random app screen. Several causes lined up, and each one is fixed.

• Web research can no longer fail silently. When the research step came back empty or with pages about the wrong thing, the script was still written, from imagination. For a news topic that meant an invented story, and every later step then illustrated it. Such an answer is now rejected and the research runs again through another route. If no research is available at all, the writer is told to stick to what is reliably known and not to invent numbers, names or events. This applies to landscape videos and Shorts.
• Every Shorts line can be pictured. The Shorts writer now builds each sentence around something that exists as real footage or a photo of the story: a named person doing or saying something, the product, the place, a post, an exact figure. Commentary that nobody can film is kept to one closing line.
• Searches name the subject. After planning, every searched clip and photo is checked against the story. A search like "celebrity news debate broadcast" is rewritten to look for coverage of this story instead, with a different real moment for each line. Your content mix, the narration and the designed cards are not touched.
• Image selection knows whose story it is. A picture that is clearly about a different person, brand or event is rejected, however well it matches generic words like "reaction posts" or "comparison chart". When nothing fits, the scene gets a generated image or a text card instead of a stranger's screenshot.
• Nothing invented on screen. Stat and headline cards only show numbers and sources the narration or the research gives them. A vague line such as "millions watched" no longer becomes a precise made-up figure with a made-up source.
• The hook is always spoken. In the Story framework the opening line was planned but never narrated, so those videos started at their second sentence. Every framework now opens on the hook.
• No crammed final beat. A plan could hold at most twelve body beats, so longer Shorts squeezed the rest of the script into the last one: a ten-second scene under a single picture. Plans now scale to the length of the Short (up to three minutes), and any beat that still runs long is split where a new picture was planned.
• A good clip is kept. When two fitting moments sat close together in the same source video, both used to be thrown away to respect the spacing rule. The earlier one is now kept and the rest of the scene is filled from other sources.

Details in Formats and Script generation.

Shorts are planned differently now, and a few things you may have noticed are fixed.

• The voice flows. A Short's narration is written as one continuous script first — the same writer, research and proofreading landscape videos get — and then cut into beats of a sentence or two (about 3–6 seconds each). Before, every beat was written on its own as a clipped line, which is why the voice sounded like a list of fragments.
• The length you asked for. Beats are timed by the words actually spoken at your voice's pace, so a 60-second request no longer comes out at 35 seconds.
• The stake line is never dropped. Some story frameworks silently lost the second line of the plan; it is narrated now.
• Text overlays sit on the right scene — they are anchored to the beat they were written for instead of a guessed time — and never overlap captions, whichever caption position you pick.
• Your chosen designs render. A hand-off between planning steps could lose most of the picked Shorts templates and replace them with a plain still. That is fixed; affected recent projects can be re-rendered on request.
• Clip fit: Auto. Widescreen footage is now framed per clip by what has to stay visible. One subject that fits a vertical crop → fullscreen, cropped toward it (a phone video sitting inside a widescreen frame fills the screen too). A close-up that needs more room → a larger middle part over a blurred backdrop. People on both sides who all matter, like a panel → the whole clip, zoomed out. Every cut inside a scene is framed on its own. Fullscreen and Blurred remain available as fixed choices.
• Transitions match the scenes around them. A transition replays the two scenes it joins, and it now uses exactly the same framing as they do, so the picture no longer jumps on the first and last frame. This also fixes vertical clips, which filled the frame in the scene but shrank into a blurred box for the length of each transition. Exports from the editor rebuild every transition from the timeline as it is at that moment, so a replaced, trimmed or deleted scene can no longer leave an old picture inside a transition.
• No frozen tails on longer beats. One clip plays for at most about 4 seconds, and beats now run 3–6 seconds. A longer beat used to keep a single clip and fill the rest with a frozen frame or a back-and-forth replay, even when a second clip had already been found for it. It now gets that second clip up front, and when none was found the search goes back for one more moment before anything is frozen.
• Your content mix is always respected in Advanced mode. A project could be created with the default mix even though scene types had been switched off in the advanced settings, because the saved mode preference and the form had fallen out of step. The video now follows the mode the form is actually in. Hiding the advanced settings still returns to Default mode, and the page now says that those choices are not used while hidden.
• Stock footage only where it fits. In Shorts the Stock Video slider is now an upper limit instead of a quota. A story about a specific person, event or product no longer gets a generic stock scene squeezed in just to fill the share. Those lines get a real clip or photo instead.
• No blink at the start or end of a clip. Clip timings are now snapped to the real cuts inside the source footage. A clip no longer opens on the last split second of the previous shot or runs a few frames into the next one.
• Text plates stay out of transitions. A plate no longer disappears half a second before a cut or pops in after one. It now appears once the transition has finished and leaves before the next one starts.
• Every shot belongs to its line. Each planned visual is now matched to the exact words it was written for, so a graphic about the axial tilt plays while the voice explains the tilt, not one beat later.
• No black scenes, no silent ending. A text overlay can no longer be chosen as a scene's main design (that rendered text on a black screen, on top of the captions), and a closing beat with nothing to say is left out instead of playing as a silent end card.
• Comparison tables fit the phone screen. Long values in the stat-comparison card shrink or wrap onto two lines inside their column instead of running off the right edge, and labels are kept to the middle third.

Details in Formats.

Several landscape videos made between September 10 and 16 looked off in the same way: generic web pictures where an animated graphic was planned, an expert on screen nobody mentioned, a world map in a video that has nothing to do with Earth. Here is what happened and what changed.

• Your designed graphics render. A hand-off between planning steps could lose most of a landscape video's planned motion graphics — kinetic titles, stat cards, dossier cards, charts. The same fix that went out for Shorts on September 17 covers landscape videos, so the graphics the plan designed are the ones you get.
• A scene that still loses its visual gets a fitting one. When a planned visual can't be produced, the replacement is now searched with the image queries the plan wrote for that scene — and if there are none, with queries written for the scene inside your video's subject and visual style. It is never searched by the narrator's sentence any more, which is how a channel name or a figure of speech used to turn into a literal picture of those words.
• Experts appear only when the story has them. For every video VidNext decides how on-camera speakers fit. When your narration names nobody who takes part in the story, the plan no longer adds interview clips of scientists or commentators just to vary the footage — it shows what the narration describes instead. When the narration does name people, interview clips are limited to those people. You can still turn Interviews off entirely in the content mix.
• Map graphics are for real places. The globe and map graphics draw the real map of Earth, so they are now used only for real, mappable places — not for other planets, fictional worlds or abstract ideas.

If one of your landscape videos from September 10–16 shows generic stills where graphics were planned, open it and choose scene assembly in the Resume/Redo menu: text and data graphics come back, and your clips and narration stay as they are. For a full rebuild of the graphics, contact support and we will re-render it for you at no charge.

Some narrated videos made since September 10 rendered captions for only part of the video: whole sections played with the voice-over but no words on screen, while the script and the narration themselves were complete. It happened when the step that draws the captions could only see some of the per-section word timings it needs.

Every section's word timings are now fetched on their own, and before the video is rendered the pipeline checks that captions were built for every narrated section. A video with captions turned on no longer ships with gaps: if a section's timings are missing, the run stops with a clear message that names the sections instead.

If one of your videos has this gap, open it and choose scene assembly in the Resume/Redo menu to rebuild and re-render it with full captions. In the editor, Resync on the captions element rebuilds the caption track from the narration before you export.

Adding a channel whose YouTube handle is written in Japanese, Korean, Cyrillic or another non-Latin script used to fail in the planner. Pasting the channel link into Add channel answered "Enter a YouTube channel URL, @handle or channel ID", and when the handle was typed by hand, Save refused it with "Use a handle like @mychannel". The same check turned such channels away in Add by URL under Competitors, and Find with AI could not read their numbers, so they sat at the bottom of the list without details.

Both forms of the address now work: the link exactly as your browser copies it (with the %E6… encoding YouTube uses for non-Latin letters) and the readable @handle. The planner reads the channel, fills the name, avatar and subscriber count as usual, and tracks it as a competitor like any other. CSV imports that list such channels keep their handle as well.

The documentation got a full visual refresh.

• Fresh screenshots on every page. All figures were re-taken from the current app — the create page, the editor, Thumb Lab, the Channel Planner, billing and the account pages — and framed the same way everywhere, with a one-line caption under each.
• Short clips where a still could not show it. Typing a topic and watching the estimate appear, switching narration modes, the music-video overlays coming alive, trying caption styles, the editor's panels and playback, the affiliate calculator — small silent loops that play in place (and hold still if your system asks for less motion).
• Diagrams for the concepts. The generation pipeline and where credits move, how a reservation is reconciled, which steps re-run after a redo, the two formats and their caption zones, the planner's states, the affiliate pool and tiers, and the ways out of a finished project.
• Pages that had no pictures now have them: the Welcome page, Email notifications, Storage and Help & Support.
• Kept honest automatically. Every figure is tied to the exact control it shows, so when a screen changes the affected screenshots are re-taken rather than left to age.

The Affiliate program page now shows two clearly different things, depending on where you are.

• Referrers see one page. Your code and link, the credits so far, the numbers and your referrals — no tabs. Under them, one card: Become an affiliate, with what approval brings and two buttons, See what you could earn and Apply now. If your last 90 days already clear a tier's bar, the card says so.
• See what you could earn, inside your account. The earnings calculator, the tier ladder, how it works and the FAQ open from that button (the public affiliate page keeps a copy for visitors). Approved affiliates find the same page under Estimate your earnings on the Program tab, opened on their own tier and split.
• The application is three short steps. About you → How you promote → Payout & rules, each checked before you move on, one submit at the end and an Application received screen. While it is reviewed, the card on your page shows what you sent.
• Approved affiliates get the hub with Overview, Links & codes, Earnings, Payouts and Program. Overview now leads with Ready to claim, your tier with the progress to the next one, and a short checklist of what your first cash payout still needs.
• Tiers and the toolkit follow approval. Every approved affiliate starts on Starter with custom codes, short links, several links and their own splits; the tiers raise the pool from 10% to 20% by themselves — and if your last 90 days already clear a bar when you are approved, you start there. Referrers who never applied keep earning 10% in credits with one link.

Details in Affiliate program.

The Ad Timing tab on the project page now verifies every proposed mid-roll break against the video itself. The generator still starts from the natural breaks in your narration, but before it returns a timestamp it looks at the frames around the cut and moves the break to the exact moment the shot changes, so an ad never lands mid-sentence or in the middle of a reveal.

Each placement now shows the kind of break it lands on — a title card, a topic change, or a pause between sentences — and a short note on what is on screen at that moment, next to the confidence badge. Older results keep their previous look. The three densities (Aggressive, Medium, Light) are unchanged. Copying got easier: the Quick Copy box has a Copy all button, and every timestamp — in that box and on each placement card — copies on its own when you click it.

Applying for cash payouts now takes a few more answers, so a person can approve it from the application alone.

• Who you are and how you promote. The form asks for your legal name, whether you apply as an individual or a company, the channels where your links will appear, one to five links to them, a few sentences on your plan and, if you like, your audience size. While the application is being reviewed, the Cash payouts card shows what you sent.
• The rules, in the form. A short list — no bidding on "VidNext" in ads, no coupon or deal sites without written approval, no self-referrals, no spam, disclose that you earn a commission — with its own checkbox next to the terms checkbox.
• Which tax form, and where. The form and the Payouts tab now explain who needs a W-9, a W-8BEN or the new W-8BEN-E (companies outside the US), link to each form on irs.gov, and say when it is needed: before your first cash payout, not for applying.
• Two cards on the Program tab. The application lives in Cash payouts; Unclaimed earnings keeps the credits-or-cash setting, and the Cash option is visible (greyed out) until your application is approved.
• Terms updated. Section 11 of the Terms of Service now asks partners to disclose that they earn a commission where required and to keep their application accurate. Partners applying for cash accept the new version.

Details in Affiliate program.

The Affiliate program page gives you more control over each link & code, and the pricing page treats a partner's referral code the way you would expect.

• Edit the split. Partners and Ambassadors can change a link's split any time (Edit split on the card). It applies to purchases from then on; the link's promo code follows the new discount, and earnings already booked keep their percent. Discount rules and the use cap stay fixed.
• The pool is yours. Your split only decides what the customer gets on a purchase the discount touches. On every other purchase — the second one after a first-purchase deal, anything after the first 30 days or after the use cap, or a purchase where a bigger VidNext discount replaced your code — you now earn the whole pool. Links limited to the first purchase or the first 30 days may therefore hand the customer the entire pool (0% for you on that purchase); each discount rule now has a ? explaining exactly this.
• Your links grow with your tier. When you move up a tier, every active link keeps its customer discount and your commission takes the pool difference — a Starter link at 10/0 becomes 20/0 as a Partner. Links promoted before today are updated the same way.
• Pick your default. Set as default on any active link makes it the one shown on the Overview tab. The default link is always live, so pause or archive another link first.
• Archive a link. Retire a link for good: the link, code and short link stop working and the card leaves the list; the referrals and earnings it brought stay yours.
• Referral codes count at checkout. A partner code without a discount (every Starter link, and any link that gives no discount) is now accepted in Have a promo code? on the pricing page: prices do not change, but the account is linked to that partner — right away when signed in, or at sign-up for visitors. Promo codes with a discount keep working exactly as before.
• Your link opens the homepage. Referral and partner links now land on the VidNext homepage with your code attached (short links already did), so the people you send see the product first instead of a bare sign-up form. Links you shared before keep working.
• Demo videos on the pricing page. The pricing page now shows the demo videos — an AI-narrated video and a music video — under the packs, so you can see what a pack buys before paying.
• Theme in the account menu. System / Light / Dark now sits in your account menu, so the pricing page and the Affiliate program hub follow your choice; the promo code pill is readable in the light theme. On short screens the account menu scrolls, so Sign Out is always reachable.

Details in Affiliate program.

Partners approved for cash already choose credits or cash on every claim. Now the Program tab also lets you say what happens to earnings you never touch.

• Decide for me, or wait. Turn Decide for me after 7 days on and pick the currency — credits or a cash payout — for anything you leave unclaimed that long. Choose Wait for my claim and nothing moves until you press a button in the Ready to claim card. The waiting period is now 7 days for both currencies (it used to be 30 for credits, and cash-default earnings joined the very next payout run).
• Links & codes shows your limit. Partner accounts hold up to 5 active links & codes at a time; the tab now says how many you are using, so you know to pause or archive one before adding another.
• The hub speaks your tier. The Overview and Links tabs now quote your own commission rate and hold period instead of the Starter numbers, and the Partner-tier line is translated in Ukrainian.
• Terms updated. Section 11.2 of the Terms of Service now describes editable splits, the pool rule and the claim choice as they work today. Partners applying for cash accept the new version.

Details in Affiliate program.

Partners approved for cash payouts no longer pick one reward mode for good. Every referred purchase now waits in the Affiliate program hub as money you claim, and you choose each time.

• Ready to claim. Once a purchase's share is past the 30-day hold, the Overview tab shows it in dollars and in credits side by side. Add as credits puts the credits on your balance immediately; Request cash puts the amount into the next monthly PayPal or Wise payout. Cash keeps its $100 minimum; credits have none.
• No made-up exchange rate. The credits figure is exactly what those purchases would have earned as credits, so both choices are worth what you earned.
• A default for what you leave. The Program tab now sets what happens to earnings you never claim: credits after a waiting period, or the monthly cash payout. Anything you already requested stays on its track.
• Refunds stay fair. A refund on a purchase you took as credits takes those credits back; a refund after a cash payout is deducted from the next one, as before.

Partners who have not applied for cash see no change: their credits keep landing automatically. Details in Affiliate program.

Partner and Ambassador in the Affiliate program are no longer about one lucky month. Each tier now has a bar measured over your last 90 days, and either of two numbers clears it.

• Two ways up. Partner opens at $150 in net revenue from your referrals' packs or 3 new paying customers in the window. A creator whose one video brought a few big buyers moves up as surely as one who brought many small ones.
• See where you stand. The Program tab shows both numbers against the next bar, and the Overview tab tells you what is still missing — "$30 in net revenue or 1 paying customer more in your last 90 days unlocks Partner".
• Checked every night, never moved down by a machine. Promotions happen automatically overnight and every active link keeps its customer discount while your commission grows by the pool difference. If a quiet stretch leaves your last 90 days below your tier's bar, your tier stays — the VidNext team may get in touch, nothing changes by itself.
• Public page. vidnext.ai/affiliates now states the same bars, so what you read before joining is what the hub measures.

Until now every visitor, anywhere in the world, had to answer the cookie wall before using the site. Where the law requires that — the European Economic Area, the United Kingdom and Switzerland — nothing changes: you still choose first, and no analytics or advertising cookie is set until you do.

Everywhere else the wall is gone. Analytics and advertising cookies are on by default, a short notice at the bottom of the page says so, and Cookie settings (in the notice, and always in the site footer) switches either category off with one click. If your browser sends the Global Privacy Control signal, advertising cookies stay off unless you turn them on yourself. Both kinds of choice are recorded with a Consent ID, as before.

The Cookie Policy and the Privacy Policy describe the new behaviour and list the advertising services that can run: Google Ads, Meta Pixel, X Ads and Reddit Ads.

In the editor, fetching a YouTube clip — Download & Use in the clip's properties panel, or a pasted link and the trim card in the media library — could stop with an unhelpful "Request failed with status code 500". The cause was almost always YouTube blocking the route the clip was fetched through, and the download gave up after a handful of tries, even though the same video was fine a minute later.

Now a clip you chose is tried on many more routes before the editor gives up, starting with the ones that just worked, and the title and thumbnail arrive without the extra wait there used to be. If YouTube is still blocking every route, the panel says so in plain words and suggests waiting a minute or picking another video; a private, removed or age-restricted video gets its own message instead of a generic error. The media library's clip search and download now show their errors too, where before they failed silently.

Refer a Friend is now the Affiliate program page in your account menu: the same link and code, with a cash option and more control over the deal.

• Credits or cash. Every account with credits keeps earning credits by default. Apply for cash payouts in the hub (payout email, a tax form, the program terms) and new commissions are paid in US dollars by PayPal once a month, from $100.
• A pool you split. Starter accounts keep 10% of every referred purchase, with no customer discount. Bring one paying referral in a month and you become a Partner with a 20% pool — create several links and codes and choose the split for each (20/0 … 5/15, in 5% steps — you always keep at least 5%). Links with a discount get a promo code, and you decide whether the discount covers the first purchase, the first 30 days or every purchase, with an optional cap on uses. Every Partner link can also have a short vidnext.ai/go/your-name address.
• A code, not only a link. Partner links that carry a discount also get a promo code — someone who never opened the link can type it at checkout, get the discount and become the referral.
• 12-month earning window. Purchases count for 12 months after each referral signs up. Referrals made before today keep the earlier rule and earn for as long as the friend keeps buying.
• Under review, explained. First payouts and anything our checks flag are looked at by a person; the status says so and nothing is lost.
• A public page with a calculator. vidnext.ai/affiliates shows what a given audience could earn, with live pack prices.

Details in Affiliate program and the Terms of Service, section 11.

When a song had no recognizable lyrics — an instrumental track, or vocals the transcription could not make out — the transcript came back as a row of music-note symbols. That row was shown as the video's Final Script, and with karaoke captions on it was burned into the video as ♪ captions for the whole song.

Music-note and cue markers such as [Music] are no longer treated as words anywhere: they are dropped from the script, from the per-section lyrics the director works with, and from the captions. A song with no recognizable lyrics now shows Instrumental — no lyrics detected on the project page instead of a script card, skips the lyrics review step (there is nothing to review), and renders without lyric captions. Songs with real lyrics are unchanged; only the note markers between lines disappear.

An existing video that already shows ♪ captions is fixed by rebuilding it from scene assembly (Redo from step 6 in the project's resume menu); the note markers are filtered when the captions are rebuilt.

Some songs came back from the lyrics transcription as nothing but music-note symbols, even though the singing was clearly there after the intro. The transcription reads the song in 30-second windows and used to carry each window's result into the next as context. When the first windows are instrumental, that context locked every later window into "music" as well, and the actual lyrics were never written down.

Song transcription no longer carries that context between windows and no longer allows the music-note symbols as an answer, so windows with singing are transcribed as words. In our test on the affected song, the same transcription that previously produced 61 note symbols now returns the full lyrics, starting where the vocals start.

If a song of yours came back with no lyrics before today, use Redo from step 1 on the project page: the song is transcribed again with the new settings, and the lyrics review, lyric captions and overlays pick the words up from there.

When VidNext is under maintenance, the create page shows a maintenance notice at the top. Until now that notice was only a warning: you could still press Generate, and projects started during maintenance sometimes failed or stalled.

From now on the notice means what it says. While maintenance is on, Generate on the create page and Start now in the planner are disabled, and a project cannot be resumed or retried. The notice tells you when to come back. Nothing you had already created is affected: videos that are rendering keep going, and your drafts, presets and finished videos stay exactly as they were.

As soon as maintenance ends the buttons come back by themselves, no page reload needed.

The create card for a music video has a new order. The brief comes first, at the top, then the song upload and the footage source. The equalizer, lyrics and title card are no longer a separate block: they are three pills in the bottom row of the box, in the place the captions pill has for narrated videos. Each pill shows ON or OFF. As soon as one is on, the live preview appears under the box with the footage dim slider above it, and it follows the pills: switch overlays on and off and you see the whole stack at once, without opening anything.

Each overlay's settings stay collapsed until you press the gear inside a pill that is on; the panel then opens under the box with that overlay's styles and options. The song title and artist now live in the Title card settings, next to the card variants they are printed on. The equalizer starts as Baseline Spectrum at 25% opacity, a thin spectrum along the bottom that lets the footage through. Review lyrics before generating is a pill in the same row as the overlays.

Music videos no longer offer captions. The lyrics overlay is how words appear on screen in a music clip, and karaoke captions built from a song transcript are gone for good. Narrated videos keep their captions pill exactly as before.

Music videos are now made in the horizontal 16:9 format only. The Shorts option is no longer offered when a song is the soundtrack, and a saved preset that combined the two comes back as horizontal.

When many videos are generated at the same time, the stock footage library's hourly search allowance can run out for the rest of the hour. Until now a project that hit this kept going without stock footage: narrated videos filled those scenes with a recovery visual (a photo, an AI-generated image or a text card), and a song video made only of stock footage could stop with an error about a blank video.

Stock footage search now pauses when the allowance is used up and resumes by itself at the next free slot, then carries on where it left off. Narrated videos wait for a while (about twenty minutes) before a scene falls back to a recovery visual. Song videos wait for the next free slot however long it takes, because moving footage is the only thing they can cut to your song. While it waits, the project page shows that the footage search is paused and when it resumes.

If a song video still gets no stock footage, it stops with a clear message, your credits are released, and the project page offers Retry stock footage, which runs the footage search again with everything else kept.

Song videos built from stock footage also make fewer searches for the same result, so more videos fit into the same hour.

The prompt check that catches text-to-video style prompts ("static camera… photorealistic… NO lightning") can now refuse them instead of only warning.

• Two behaviours. By default the dialog still explains what VidNext makes, offers three topic prompts on your theme, and lets you Create anyway. Under the stricter setting there is no Create anyway: Generate stays off for that exact text until you rewrite the prompt or pick a suggestion.
• Same for music-video briefs. A brief that would need generated footage gets the same treatment, and Continue with real footage is not offered on the project page; Make a narrated video instead still is.
• Nothing is charged. A refused prompt never starts a project, so no credits are held or spent.

See the prompt check and music video briefs.

A date or name overlay could appear noticeably late, about ten seconds after the narrator had already said it. This happened when the plan attached the overlay to the scene that follows the sentence with the date, and timed it to a phrase from that later scene.

Overlays are now synced to the words they show. When the narrator says "1999" or a person's name, an overlay carrying that year or name appears at that moment, on the scene that speaks it, even if the plan had placed it on the neighbouring scene. If the words are spoken in the closing second of the previous scene, the overlay opens with its own scene instead of arriving late. Overlays whose words are not spoken keep their planned timing.

Two smaller timing gaps are fixed with it: overlays anchored to phrases with an apostrophe or a hyphen ("Bell's inequalities", "dead-end server") were silently falling back to their estimated time, and now sync correctly.

Videos generated before this fix keep their existing timing; re-rendering a project from the editor applies the new sync.

When a script joined two words with a dash and no spaces around it (for example "superposition"—a coherent blur), the voice could say the word before the dash twice at the point where one scene ends and the next begins — or, less often, skip it. The rest of the narration was unaffected, and the Final script shown on the project page was always correct; only the spoken audio was off.

The cause was in how the script is cut into scenes. When a cut landed inside such a dash-joined pair, the two neighbouring scenes could both claim the same word, and the narration is recorded from the scenes joined together. Scene narration is now cut so that every word of your script is spoken exactly once, in order — a custom script is read verbatim, as described in Scripts.

Videos generated before this fix are not changed automatically. If one of yours has a repeated word, edit that scene's narration text in the editor and use Regenerate Narration.

A text overlay with a longer label could be cut off at the edge of its box, so a caption like "DEAD-END SERVER │ COSMIC LOCAL QUARANTINE" ended mid-letter. Changing the overlay's style or position, or regenerating the scene, usually made it fit, which is why it looked random.

The cause was how the overlay decided its text size. Each style shrinks its text to fit the space it has, but the check that decides whether a line still fits accepted a size where the text would only fit on two lines, while the design keeps it on one, and the tracked spacing between letters in the small label was not counted at all. The overlay then drew a line wider than its box and the box hid the ends.

Now every one-line style measures its line exactly, spacing included: a label keeps one line as long as it stays readable, and wraps onto a second line otherwise. The two-part "label │ value" style also gets the wide layout it needs, so a real pair of phrases fits on one line at a comfortable size.

Videos generated before this fix keep their existing overlays. Re-rendering a project from the editor picks up the new layout.

The Refer a Friend page no longer asks for a subscription. It unlocks as soon as your balance has credits — from a top-up, a gift, or referral earnings — and your share link is ready to copy.

• Your link remembers your friend for 30 days. Whoever opens it has 30 days to sign up on that device and still count as your referral, even if they close the tab and come back later.
• Every pack counts. You earn 10% of the credits from each pack a referred user buys, for as long as they keep buying — not just the first one.
• Rewards land after a 30-day hold. Each reward shows as Pending until then; that pause is the fraud check.

Details in Billing FAQ.

When many videos were being generated at the same time, the stock footage search could use up its hourly allowance. Until now every stock search then failed at once for up to an hour, so scenes that should have used stock footage fell back to a recovery visual (a photo, an AI-generated image or a text card) even though the allowance frees up gradually.

Stock footage searches now wait briefly for the next free slot when the allowance is used up, spread their requests evenly over the available capacity, and carry on by themselves as soon as it frees. A search only falls back to a recovery visual when no slot opens within a couple of minutes.

In the editor, the stock footage search box now tells you it is busy and to try again in a moment instead of showing an error.

Videos generated before this fix are not changed. If one of yours has a scene that fell back to a recovery visual, redo that scene as described in Redo and regenerate.

Opening a finished video in the editor could show a few empty stretches on the timeline, even though the rendered video had a picture there. Exporting from the editor would then paint those stretches black.

This happened to scenes whose planned image could not be produced and which VidNext rescued during assembly with a searched image instead. The rescue reached the rendered video but not the data the editor builds its timeline from, so the editor left a hole where the video had an image.

The editor now mirrors the rendered composition for any scene its usual sources do not cover, and an existing timeline is repaired the next time you open it: the missing scenes are added in place without touching your edits. A scene you deleted on purpose stays deleted, and a clip you placed in that spot yourself is kept.

Scene transitions in the editor are also brought in line with the rendered video. When a clip is shorter than its scene, the video holds the clip's last frame until the scene ends; the transition into the next scene now holds that same frame instead of briefly showing footage from past the end of the clip. Existing timelines pick this up on their next open as well.

Sometimes the final render stops because the finished scenes point at footage files that are no longer available — a stock clip or a downloaded scene that did not make it into storage. Until now that read as a generic failure, and retrying the render alone could never fix it, because the render kept looking for the same missing files.

Now the project page says what happened in plain words and offers Rebuild scenes in the error box. One press fetches the missing footage again and renders the video, with every setting and every other asset kept — nothing to re-upload. Where a video stopped this way in the past, use Redo → Scene assembly (or Clip download when the message names clips) from the Resume/Redo menu.

Two smaller things on the same page: when a music video stopped because its brief asked for generated footage, the explanation is now shown in your language, and Start a narrated video carries the brief to the create page as the narration script — as the message promises — instead of as the topic.

The promo code field on the pricing page chooses its words more carefully.

• "Beats" means beats. When a code matches the discount you already had — say a 40% code on an account that already carries 40% — the field now simply confirms the code is applied. It no longer claims the code beat a discount of the same size.
• Nothing about your price changed. The discount you get is the same as before; only the message is more accurate.

Discounts still never stack — see Pricing & plans.

When a lot of videos are generating at the same time, new projects now wait their turn instead of all slowing each other down.

• You see where you are. A waiting project shows Queued · #N on its dashboard card and on the project page — N is its place in line. Hover the chip for a rough wait time.
• Nothing to do. Projects start automatically in first-come order, and the number counts down as the ones ahead finish. Once yours starts, everything is as before: live step progress, then the finished video.
• Credits are unchanged. Waiting costs nothing extra; the estimate you saw when you pressed Generate still applies.

See what each status means in the pipeline overview.

When Music video mode stops because your brief asks for footage that would have to be generated, the message used to point at a Create anyway button on the create page — which is only there when the create page raised the same warning itself. If it hadn't, there was nothing to press.

Now the failed project carries the two real choices. Continue with real footage re-plans the same project from film scenes and stock clips, keeping every setting you chose. Start a narrated video takes your brief to the create page as a narrated video, where your text becomes the script. Both sit in the error box on the project page.

The Brief check on the create page offers the same way out: in Music video mode it now suggests Make a narrated video instead alongside editing the brief and continuing with real footage. A brief that describes a story rather than footage to look for is usually a narrated video in the wrong mode — the song stays as background music and your text becomes the narration.

If the file you uploaded sounds like a spoken voiceover rather than a song, the stop message says that first, instead of only discussing the brief.

Entering a promo code on the pricing page now always tells you where you stand — and it never leaves the field spinning.

• A better code takes over. If the code beats a discount you already had, we swap it in and say so.
• A smaller code is kept, not lost. Your bigger discount stays and the field tells you which one is applying.
• A code that doesn't apply says why. Expired, or first-top-up-only after your first top-up: you get the reason, plus a reminder of the discount you still have.
• It no longer blocks your purchase. A code you've outgrown used to stop checkout; now the purchase goes through at your best available price.

Discounts still never stack — see Pricing & plans.

When a project failed at the final video step because a piece of stock footage had gone missing, redoing Scene assembly would fail again the same way, every time. Now it repairs the missing footage — putting the clip back if it can, or finding a replacement — and the final video goes through.

Very occasionally a stock clip that VidNext had already picked for a scene did not finish being stored, and the failure only surfaced at the last step. The Scene assembly redo could not tell the difference between a clip that was genuinely available and one that was not, so it kept handing the same broken scene to the final render. It now checks properly and fixes what it finds.

Two related improvements: a scene whose stock clip cannot be stored is now kept on its regular footage instead of being set up to fail later, and a brief storage hiccup during the final step is retried rather than being reported as missing footage.

If your browser can't reach VidNext's servers, the app now says so and offers a Retry button, instead of behaving as though your account had lost access.

An ad blocker, VPN or company network filter can quietly drop the app's requests. When that happened the app used to fall back on "no access", which looked like an account problem and could not be fixed from our side. Now it stays where you are, tells you what is actually wrong, and recovers the moment the connection works again — nothing about your account or credits is affected.

The waitlist is gone. Sign up creates your account straight away — no invite code, no approval to wait for. Every Get started button across the site now goes directly to sign-up.

Old waitlist links still work: they take you to the sign-up page instead. If you were sent an invite or referral code, it still applies — paste it or use your link as before and the bonus is credited once you're in.

Support answers now point at the control instead of just describing it. When the VidNext Assistant (or a person from the team) mentions a setting, the reply carries a Show me chip next to it — click the chip and the app opens the right page, scrolls to that control and highlights it, opening a collapsed section first when needed. The help panel folds into a Back to chat pill while you look, so one more click brings the conversation back with your draft intact. If the control isn't on the page yet — a step that only appears after you enter a topic or pick a voice, for example — the guide points at what to do first and continues to the control on its own once it appears. The same Show me in the app links now sit next to the steps in these docs, and the ones you receive by email open the app in the same way. Details in Help & Support.

You can now talk to the VidNext team without leaving the app. A life-buoy Help button next to the notification bell opens a side panel on every page with your conversations, a new-conversation form and the chat itself. Get help on a video that stopped (dashboard card, project page, auto view) opens the form with that video, its status and its error already attached, and the Page & browser details card shows exactly what else we attach so you can untick it. Paste screenshots straight into a message, or drop files — up to 5 per message, 25 MB each.

Right after you send, the VidNext Assistant may answer — built from these docs, clearly labelled as automatic, with This solved it and I still need a person underneath — and you can keep chatting with it. When the docs don't cover a question it asks whether to pass it to a person instead of guessing — Yes, connect me with a person alerts the team, No, that's all closes it quietly — pricing and credit questions are answered from the docs, and anything about your own payments, refunds or account goes straight to the team. The header shows whether the team is online and how fast it usually replies, replies appear in the open conversation within a second or two, and you can turn on browser notifications for them. Replies from the team arrive in the app (bell notification, a New marker in the thread) and by email, and answering the email lands your reply in the conversation. Conversations move through Open, Waiting for you, Resolved and Closed; mark one as solved yourself, rate it once it is resolved, or reopen it by replying. A new Support replies toggle in Settings → Notifications controls the emails. Everything is described in Help & Support.

New to VidNext? A Watch the demo button now sits under the heading of the create page and next to the YouTube guide on the dashboard. It plays a short walkthrough of making a video, and switches to a music-video demo when your narration source is Music video. The button appears once a demo is available for that mode.

The slider on the pricing page now ends in a grab handle with left/right arrows, and a Drag to size your pack hint sits under it until you move it for the first time. Drag the handle (or use the arrow keys once it has focus) to move between the twelve packs; the video count, price, and per-minute rate update as you go. See Pricing & plans.

Partner promo codes now work on the pricing page: a code you arrive with through a partner's link is picked up automatically, or enter one under Have a promo code? at the bottom of the page. The discounted prices are shown before you pay, the discount is applied at checkout, and the payment confirmation names the code. Most codes are for your first top-up only, and discounts never stack — you always get the best one. See Pricing & plans.

Story images now always appear as your own picture in the image design you chose for it (or one of your enabled image designs). Previously an image the script did not reach could be added as an extra scene with an empty or unrelated design, and every later scene could drift out of sync with the narration. Images the script does not place itself now take over an existing scene near their intended point, so the video keeps its length, its order and its narration.

Web image selection now judges every candidate against the video's topic, era, region and language, not only the scene's keywords: a picture that contradicts them (a different country's living room in a village documentary, a modern colour photo in an archival story) is not used. Watermarked pictures, stills with burned-in captions in another language or a broadcaster logo, and photos of famine, war dead or disasters used to decorate an unrelated point are rejected as well. Cards that name a specific person, brand or product model search for that exact name.

Music videos can now carry the graphics that make a clip look produced, all chosen in the music video card before you generate:

• Equalizer — 24 visualizer styles in four families (bars, rings, waveforms, fields) that move with the real beat and spectrum of your song. Choose the placement, a colour mode (style-pack accent, custom colour, white or a spectral ramp), intensity and glow.
• Lyrics — 21 big, animated karaoke styles synced to the singing word by word. The kinetic family (Kinetic Cascade, Kinetic Slam, Typo Poster) builds each line word by word with a different move per word and sizes that follow the singing; Infinite Depth keeps lines travelling away into a tunnel; Drift & Fade and Split Reveal are cinematic reveals; the karaoke pill, highlight sweep, gradient, neon, glitch, typewriter and knockout looks complete the set. Size, font and colour are yours, the position follows the style unless you pin it; lyrics replace captions for that video. The transcript is cleaned up and re-synced to the song before rendering — review the lyrics first for the best result.
• Title card — the song title and artist animated before the first line, in four variants.
• Live preview — the Overlays block previews every enabled overlay together over sample footage while you adjust it, with pause, restart, scrub and a larger view; the equalizer gained an opacity control.
• Editor — the equalizer and lyrics elements are editable on the Overlays track: style, colours, placement, intensity, glow and opacity for the equalizer; style, position, size, font and a Lyrics & timing list for the lyrics, where every word or line and its start and end can be corrected, split, joined or set at the playhead. Exports use the edited words and times.
• Lyrics review — music videos now review their lyrics in a dedicated dialog: timed lines with their moment in the song, in-place edits, Enter to split and Backspace to join lines. Edited words are re-synced to the singing before rendering; there is no proofreading step for lyrics.
• Vary by song section switches an overlay to a companion style on the chorus and calms it down on the intro and outro; the Footage dim slider keeps everything legible over bright clips.
• Every style has a live preview over sample footage in your format, and all three overlays are editable later in the editor (style, colours, placement, dim).

Details in Music videos.

Map scenes that showed two places close to each other — a route between neighbouring towns, a cluster of ports, a city with its suburbs — could print both names on top of each other, leaving an unreadable smudge next to the marker. Fixed:

• Names never overprint. Every map style (routes, spread, zoom with secondary markers, arrows) now places each name in a free spot around its marker: the other side, above, below, or one row further out with a thin pointer line back to the marker. Names that were already legible stay exactly where they were.
• Short routes get a real map. A route whose stops sit a few dozen kilometres apart used to be framed like an intercontinental flight, with both markers on one spot. The camera now dives to the scale of the route, so a county-level evacuation route shows two distinct markers and the path between them.

The footage check introduced on 2 September was rejecting far more than it should and then searching for hours to replace what it had thrown away. Fixed:

• Any moment of the named film counts. A clip from the right film at a different moment than the one described is no longer marked "off-target" — it is kept, and only footage from a different film, a presenter or reaction video, a live performance, a themed event or a collage is rejected.
• One search pass per scene, not one per keyword. A scene that still needs footage after the check is re-searched once, from its film's own moments first, in a single analysis instead of dozens; a second pass runs only when the first found nothing usable. The whole re-search is capped per video, so a long song can no longer spend an hour on it.
• Scenes with enough footage are left alone. A scene that already has enough distinct footage but fewer clips than beats replays its clips and borrows footage the viewer has not seen yet instead of being re-searched.
• Stock sections are never re-searched on YouTube. They are chosen by the stock library's own scoring and were previously sent through the film re-search for nothing.
• Fewer junk sources up front. Search results for film scenes are now screened by title and thumbnail before the footage is analyzed, so nature clips found for "Before Sunrise" or a dance show found for "Moulin Rouge" no longer reach the check in the first place.

Music videos of the same length are back to their previous cost and generation time. Details in Music videos.

Music videos were repeating the same shots — a five-beat scene often had only two clips to cut from — and the stock check could take longer than the rest of the video. Changed:

• One distinct shot per beat. Scenes from the same film (or the same passage of the song) now share one footage pool, and VidNext asks for as many different moments as the scene has beats, plus spares. A film whose pool still comes up short is searched once more from its generic scenes.
• Cuts on real shot boundaries. Each clip's shot changes are detected in the footage itself, so a beat never starts mid-shot and a long clip with several shots gives several cuts. Consecutive beats alternate between wide, medium and close shots when the footage allows.
• Same film first. A scene that runs short borrows unseen footage from its own film before anything else; borrowing from other scenes is the last step, and the quality report now flags any scene where footage still had to repeat.
• Sharper source picks. Uploads below 720p are skipped while HD ones are available, clips with a visible channel watermark are kept only as a last resort, and long uploads are sampled rather than watched end to end.
• Faster stock check, for every video. Stock clips are screened from their preview frames first; only the best few per scene are downloaded and scored as video. This applies to stock sections in narrated videos as well, including videos with a single stock scene.
• Cuts on the downbeat. The song's bars are now read from the audio (accents and chord changes decide where beat one is), so cuts land on bar lines instead of on whichever beat a section happened to start on — the montage finally moves with the music rather than merely at its tempo.
• Song language is auto-detected. The language pill in Music video mode now reads Auto-detect by default and is labelled as the song's language. It used to carry the app language silently, and a wrong forced language made the lyrics come back translated instead of transcribed.
• Fewer re-searches. A clip with a channel watermark or a shot from another film in your brief no longer counts as missing footage, so scenes that already had usable material are not re-searched.

Details in Music videos.

Starting a video from a planner idea could occasionally create the same project twice — a double-click on Start now, or a second click on a card that hadn't refreshed yet, each started a full generation, and the extra one later appeared as a duplicate card. Fixed:

• One start per idea. Once an idea has a video in progress, a second start is refused with a clear message and the board refreshes to show the running project. The idea can be started again after that video has finished, failed or been cancelled.
• The Start button tells the truth. An idea whose video is still in progress shows the progress link instead of a Start button — in every column, including an idea moved back to Ideas while its video is generating.
• Faster feedback. Start now shows its spinner on the first click, and the dialog can't be dismissed while the start is in flight.
• No stranded credits. If a start fails before generation actually begins, the held credits are released immediately.

This applies to every way of starting a planned idea: the board card, the idea details, the calendar's Needs video panel and Customize settings. Details in Channel Planner.

Footage search can now locate the moment a scene needs by reading a candidate video's transcript first and only then looking at the frames around it, instead of scanning every candidate from start to finish. Each scene group is checked against up to ten candidate videos at once, and sources up to 60 minutes long are accepted, so full episodes and long documentaries are no longer skipped in favour of short compilations.

Every clip that reaches your project was visually verified: a candidate that was only described in speech but never looked at is rejected and searched again. The change is rolling out gradually to accounts; nothing changes in how you create a video.

Music video mode now vets every downloaded clip before it reaches the edit. Footage that is a still image or poster, a title card or credits, near-black, or clearly from a different film than the one requested is rejected, and the scene searches again — first the film's own moments, then the other films in your brief, then cinematic footage matching the song's mood and motif. Clips with burned-in subtitles or a sub-HD source are kept only as a scene's own last resort when nothing cleaner exists.

Within a scene the same seconds of a clip are no longer shown twice: a scene with fewer clips than beats now plays different moments of its clips instead of repeating one window, and a scene that ends up with no footage of its own borrows clips the viewer has not seen yet before repeating anything. Search results from auto-generated "Topic" music channels, trailers, soundtrack uploads and fan tributes are skipped up front.

The create step for music videos is simpler. After choosing Music video, the card asks for the song first, then where the footage should come from — Mixed (VidNext decides per section, the default), One film or show, Several films, Stock & abstract or Nature — and then an optional brief. One line says what you get: real clips cut on the beat, not an AI-generated video. Controls that don't apply to a music video are gone from this mode: the script and attachment menu, sound effects and the narration starters. The script-review switch now reads Review lyrics before generating, since the transcribed lyrics are what it reviews.

Scene transitions in music videos now follow the song: entering a chorus or the outro always gets a transition, other section changes usually do, and beat cuts inside a section stay hard cuts. Music videos start with a Music video transition set (dissolve, fade to black, film burn, flash, impact cut, whip pan), also available as a preset under Scene Transitions. The footage choice presets the Content Mix (Video Clips from YouTube versus Stock Video from the stock library) and the Footage Sources list, both still adjustable, and narration features such as chapter headings, text overlays and calls to action no longer appear in this mode.

Stock footage selection is working again. Since mid-August the stock library clips were found but never scored, so every stock scene quietly fell back to an image or, in a stock-only music video, to a black frame. Music videos now also keep moving footage as a last resort when no clip clears the relevance bar, a video with nothing on screen is refused before rendering instead of being delivered, and the final check samples frames so an all-black render can no longer pass.

Music video mode and Upload my narration find real footage — film and show scenes, stock clips — and cut it to your audio; they can't render animation, invented characters or camera directions. If your brief reads like a prompt for a video generator, the Brief check now explains this before the video starts and offers three real-footage briefs on the same idea, in your language. Pick one, edit your brief, or Create anyway to continue with real footage. A brief that can only be fulfilled with generated footage no longer turns into a video by accident: without that confirmation the project stops before generation and no credits are charged.

Narration uploads are now checked for what they sound like. A music track uploaded in Upload my narration, or a spoken recording uploaded in Music video mode, gets a note next to the file and a dialog on Generate that offers to switch to the matching mode with the file kept. The check is a heuristic — singing, poetry or narration over a music bed can fool it — so your choice always wins.

In Music video mode the Story images and Video reference attachments are hidden: the song and the brief drive the visuals, and neither attachment was used there.

The narration Speed setting (Slowest–Fastest) is now part of your saved presets. Save a preset with a speed picked and it comes back exactly when you apply the preset — including planner one-click videos that use it. Previously the selector reset to Normal every time, and changing the speed now also counts as a preset change you can save back.

Generating a video straight from a planner idea no longer needs any preset setup. Before, one-click generation required the preset's prompt to contain a {{Main Idea}} placeholder — without it the start was blocked, and an idea's description and brief never made it into the video. Now it just works:

• Any preset, zero setup. The first time a preset is used from the planner, VidNext prepares a planner version of its prompt automatically — the topic becomes a placeholder, every style and format instruction stays intact. From then on each planned idea slots straight in.
• The whole idea travels. The prompt now carries the idea's full details — for AI ideas the brief (hook, angle, key points, must-cover facts, keywords), and for manually added ideas their description, notes, tags and thumbnail idea. Manually written ideas used to contribute nothing but a title.
• See it before you start. The generate dialog gained a What we'll generate preview showing the exact prompt, and Customize settings opens the create page with that prompt filled in and editable — the video uses exactly the text you leave there.
• You stay in control. The channel editor's new Planner prompt panel (under the preset picker) lets you view, edit or reset the planner version of the prompt. Your saved preset prompt is untouched — on the create page it loads exactly as you wrote it.

Placeholders are now optional everywhere, and {{Title}} and {{Description}} join {{Main Idea}} for exact placement when you want it. Details in Channel Planner.

Fixed a bug where auto-generated thumbnails could render your whole video brief as the on-image headline instead of a short punchy title — and ignore a detailed thumbnail instruction you'd written. Detailed instructions with exact headline text (colors, layout, a reference photo) now carry through reliably, and headlines always stay short even when generation has to fall back to a simpler layout.

Thumbnail headlines now also follow your video's output language automatically. If your video is set to Ukrainian, Japanese, or any other language, the generated headline and its variants come out in that language — you don't need to ask for it. An instruction that specifies exact headline text is still used verbatim, in whatever language you wrote it.

Thumbnail instructions now also work everywhere, not just in Thumb Lab:

• The project Thumbnails tab gained the "Use my prompt exactly" switch and, with references attached, the extra-web-references switch — the same controls the Thumb Lab composer has.
• The create page's thumbnail popover now takes a custom instruction, up to 3 reference images, and "Use my prompt exactly" — applied when thumbnails generate automatically with the video.

Details in Thumbnails.

The Video Stats Monitor now records a point only when a video actually moves — views, likes, comments, the title or the thumbnail. A flat stretch has no points in it and the chart draws straight through, so the history reads as changes instead of as a wall of identical measurements. Nothing you could see before is gone: every video's current numbers, its title and thumbnail history, and the views-per-hour ranking are unchanged.

The last 30 days keep every collection. Beyond that, each video keeps one point per day — enough to see a trend on an older video without carrying hourly detail forever.

Videos tracked on the Stats page was counting measurements rather than videos, so the number was far larger than the number of channels and videos you actually follow. It now counts tracked videos, which is why it drops sharply.

Collection is smarter about where the detail matters, too: the newest videos on each channel keep being checked around the clock — the launch window where hourly numbers are worth watching — while the full channel refreshes once a day. If you're away from VidNext for a couple of weeks, collection eases off for your channels and catches itself up the moment you're back; Collect now still runs a full refresh on the spot.

Details in Channel Planner.

New ideas on the Board, the channel page and the calendar's ⋯ menu now opens the Idea Studio. Pick a channel, how many ideas you want (the studio suggests a number from the empty slots in your calendar), the horizon, the evergreen-to-trending mix, the formats, and what the AI should look at — your channel's performance, your competitors' latest uploads, live YouTube trends, and web research with the seasonal calendar. The credit estimate is on the button before you run, the run works in the background with live progress, and the bell tells you when the ideas are ready.

Every idea comes as a card with a hook, format and length, a "why now" backed by links, a score, a publish-by date when it's timely, tags, a thumbnail concept and alternative titles. Approve, Decline (with a reason) or save it for Later — with A, D and L on the keyboard — approve everything above a score in one click, and Generate more continues from your decisions; a declined idea never comes back for that channel. Then Add to Ideas, or Add & schedule to drop them straight into the channel's empty slots, timely ones before their date. Each idea keeps its brief, and it's used when you generate the video.

Switch on Automatic ideas per channel — daily or weekly — and fresh ideas arrive as pending, with a Review count on the New ideas button, on the channel Overview and in the Monday summary. The old Add idea and Bulk add forms live on as the Write one and Paste a list tabs.

Details in Channel Planner.

The Planner tab is now a calendar across all your channels: every scheduled post is an event in its channel's colour, Month, Week and Agenda views, and a panel on every post with the posting checklist, the YouTube kit and Mark posted. Drag a post to another day to reschedule it, and drag finished videos from the Ready tray onto a day to schedule them. Posting times are set in each channel's own timezone and shown in yours.

Add channel is a full channel editor: paste a YouTube URL or @handle and Analyze with AI to fill the profile — audience, countries, content pillars, keywords, goals and the AI's insights — then Find with AI to add competitor channels. Every competitor gets a role relative to your channel — Rival (with a 0–100 beatability score: Beatable, Catchable or Uphill), Learn from or Benchmark — and the panel opens with a Your position line so you tick channels you can actually race, not the giants you should only study. Brand-new channels are covered too: with no public subscriber count the planner assumes 0, the rival band always reaches 10,000 subscribers, and the posting timezone follows the audience (a US audience gets New York, not your browser's zone). Tracked competitors are checked the moment you add them, so Recent uploads from tracked competitors on the channel's Overview fills within seconds (and has a Refresh button plus the time of the last check). Each channel page now has Overview, Calendar, Videos, Pipeline and Profile tabs. The editor's Basics section also holds a Default thumbnail style next to the default video preset: projects created from the channel start with that style (leave it empty to pick one per project).

You can also get the plan by email: Email me my plan in the toolbar (or Settings → Notifications → Planner emails) switches on a daily posting plan at the time you choose, a reminder an hour before each post and a Monday weekly summary — and Subscribe gives you a private link that puts your posting plan into Google, Apple or Outlook Calendar.

Details in Channel Planner.

Retry on a variant in a project's Thumbnails tab now re-uses the reference images you attached when you generated it. Before, a retry re-rolled the variant from the style alone and quietly dropped your references, so the new image often looked nothing like the one you were trying to improve. It also keeps the original run's choice about whether to search the web for extra references, so a retry reproduces the same recipe. Details in Thumbnails.

Thumbnail generation is also quicker to fall back when Google's servers are busy: one step on the thumbnail path could previously spend up to five minutes retrying before moving on, and now gives up after 30 seconds like every other step.

The Storage page (account menu → Storage) now works on several projects in one go. Tick the rows you want — or use Select all, Failed, Older than N days and All but newest N — and the bar under the list shows the count and size of your selection with Clean up selected and Delete selected. Both ask once, then show a progress counter and end with a summary of how much space was freed. Details in Storage.

The batch clean-up that used to live in the dashboard's storage panel has moved here and been reworked along the way: large batches no longer fail part-way with a network error, deleting a project that was still generating now refunds the credits held for it, and the storage figures update right after a clean-up instead of a few minutes later.

Deleting a project from the dashboard now removes its card the moment you confirm. Previously the card stayed in place until all of the project's files had been cleaned up — for a large project that could take half a minute — so it looked like nothing had happened, and trying again on a project that was already gone showed "Failed to delete project".

Now a repeat attempt on an already-deleted project simply completes, and your storage usage updates as soon as the project is removed. If a delete does fail, the card comes back and the message says so.

VidNext makes narrated videos about a topic — it doesn't render a shot you describe, the way text-to-video tools like Veo or Sora do. Prompts written for those tools ("static cinematic shot of a cabin in heavy rain, photorealistic, NO camera movement…") used to go straight into production and come back as something their author never wanted.

Now the create page catches this before generation. If your prompt reads like a shot description, a Prompt check dialog explains what VidNext actually produces and offers three topic prompts built from your own idea — one click puts one into the prompt field, and the quick-start format chips are there too. Prefer your original text? Create anyway continues exactly as typed; the check never blocks you.

A live hint under the prompt field also appears as soon as your text looks like a text-to-video prompt, so you don't have to reach the Create button to find out.

Details in Scripts & prompting.

The Terms, Refund Policy, Privacy and Contacts links at the bottom of the account menu now open the page in the language you're using: the English interface links to the English documents, the Ukrainian interface to the Ukrainian ones. Previously the English menu could open the Ukrainian pages.

The editor now has a lite mode for slower laptops. It switches off the parts of the timeline that cost the most on weak hardware — clip thumbnails, audio waveforms, and the preview's pre-loading of upcoming clips — while editing, playback and export stay exactly the same.

The editor decides by itself: a quick speed check when it opens gives fast machines the full editor immediately, slow ones lite mode (a yellow notice with a Turn off button appears), and in-between machines start light and switch to the full editor once the page is running smoothly. A page that stalls for several seconds drops back to lite mode and remembers it. You can also toggle it any time with the ⚡ button at the right end of the playback bar. Your choice is remembered on that device. If a computer can't open the editor at all, add ?lite=1 to the project's editor address once. Details in Editor tour.

Also in this update: scrubbing the timeline while zoomed far out no longer makes the preview reload dozens of clips per step, thumbnails are loaded only for clips that stay in view, and the editor logs [perf:editor] diagnostics in the browser console when something is slow — if the editor ever feels stuck, a screenshot of that console helps us fix it.

Every email VidNext sends about your videos and credits is now a toggle — including two new ones: a reminder when a finished video sits undownloaded for a couple of days, and a short guides series after your first export (prompting, presets, footage control, Shorts, voice and music, redo, the publishing kit). Open the account menu and pick Notifications (or use Notification Settings at the bottom of the bell menu) to switch Video updates, Problems, Credit balance and Tips & product news on or off individually, or pause them all with one master switch — changes save instantly. Purchase receipts, refund confirmations and team invitations are always sent.

The emails themselves now name a video by its generated YouTube title, or simply say "your video" when no title has been generated yet, instead of repeating your full prompt — the Open the project button takes you to the right place either way. Service notices link to your notification settings in the footer; the Unsubscribe link in tips and news emails keeps doing exactly that. Details in Email notifications.

The editor's bottom timeline panel now has a drag handle. Grab the thin divider just above the playback controls and pull it up to see more tracks at once, or down to show as few as two tracks and give the video preview more space. Double-click the divider to go back to the default six-track height; with the divider focused, ↑/↓ steps the height one track at a time. Your chosen height is remembered on that device. See Timeline editing.

The editor is also noticeably lighter on long projects: audio waveforms are now built once per file instead of once per narration segment, clip thumbnails are loaded only for the part of the timeline you're looking at, and the preview does far less work per frame — long videos that used to make the tab unresponsive on slower machines open and scrub smoothly.

vidnext.ai/blog has grown into a full library over the last two months, across four areas:

• Tutorials — your first video step by step, prompts that produce good scripts, fixing one scene without regenerating the whole video, choosing a narration voice, copying a creator's writing rhythm with a video reference, and the thumbnail formula.
• Niches — true crime and history production, finance and tech, and the seven faceless niches worth starting in 2026.
• AI & YouTube — Shorts against long-form, where mid-roll ads belong, running three channels without burning out, and getting monetized before the 1 February 2027 threshold change.
• Product — what a 10-minute video actually costs, itemised credit by credit.

Several of them are built on one video we timed end to end and billed line by line, so the numbers are measured rather than estimated — including where every one of its 442 credits went.

The counters under the dashboard welcome title now tell the whole story. In Progress counts only projects that are actually generating or rendering — your parked drafts have their own Draft counter. Failed counts real failures only; projects you cancelled yourself are listed under a new Cancelled counter instead of being mixed in with errors.

All six counters still work as one-click filters for the project list, and the compact filter row next to the search box has matching buttons for drafts (✎) and cancelled projects (⊘).

Every finished thumbnail variant — in Thumb Lab and on a project's Thumbnails tab — now has an Edit button. Describe the change in plain words ("make the headline bigger and yellow", "remove the person on the right"), optionally attach up to 2 reference images, and confirm. Your words are sent to the image model exactly as written.

The result appears as a new variant right next to the original, marked Edit, with the usual download / analyze / save-style actions — and it can be edited again. The original stays as it was. It costs the same as generating one variant, and the price is shown before you confirm; on the project tab, Retry on an edited variant re-runs that same edit.

Project thumbnails opened from Thumb Lab's Recents also gained a Use in project button, so you can pick the project's thumbnail — including a freshly edited one — without leaving Thumb Lab. And Generate more there no longer runs blind: it opens a dialog prefilled with the latest variant's final prompt and reference images, where you can tweak the prompt, swap references and choose how many variants to make.

If you use the Channel Planner, the Recents panel in Thumb Lab now has a channel selector at the top: see only the thumbnails made for one channel, or all of them. While a channel is selected, new thumbnails you create are saved to it — the composer shows the channel as a chip you can clear.

Project cards on the dashboard and in the Channel Planner now show a still frame of the finished video (or your generated thumbnail) instead of loading a video player for every card. Nothing streams in the background any more: on desktop, hovering a card previews the footage as before; on phones and tablets a tap takes you straight to the project.

The page also stops refreshing while its tab is in the background and picks up again the moment you return, and the moving ambient lights on the dashboard stay still on touch devices. Phones that used to run warm on the dashboard should feel noticeably cooler and faster.

The Cartesia voice picker's Common Library now includes SONA PRO — a curated collection of premium studio voices, currently spanning Ukrainian, English, Spanish, Japanese, Polish, Russian, and Norwegian. They carry an amber PRO badge, lead the list in their language, and a new PRO toggle next to the Male/Female filter shows only them (with a count for the selected language). Play previews the voice's official sample.

PRO voices work everywhere the Cartesia picker does — the create page, favorites, and the editor's narration voice modal — and narration with them is billed at the normal Cartesia rate. See Narration voices for the full picker walkthrough.

Cards on the dashboard used to show a frame grabbed from the video and the raw prompt you typed — which made a page of projects hard to scan once you had more than a handful.

Now, if a project has generated thumbnails, its card uses one as the cover. The variant you Mark as final wins; otherwise it's the first variant. Hovering the card still previews the video underneath, as before.

And if you generated a YouTube title for the project, the card shows that title instead of the prompt. Your original prompt is still there as the tooltip when you hover the text.

Projects without thumbnails or without a generated title look exactly as they did.

Opening your project list used to quietly start streaming every card's finished video in the background, just to draw the preview thumbnail. On a full page of projects that was several gigabytes of traffic competing for your connection — so pressing Download on a card sat on a spinner for a long time before the browser showed anything.

Cards now load only the little bit of the video they actually need for the preview, and the download button hands the file over the same fast path the project page already used. Downloads from the dashboard now start about as quickly as they do from the project page.

The runtime shown on each card is also now the exact length of the finished file, read once when the video is rendered, instead of being measured in your browser every time the page loads.

Saving your current settings as a preset used to leave the My Presets menu showing the list it had loaded when the page opened — your brand-new preset only appeared after a page reload, and even then it sat at the bottom of All presets rather than near the top.

A preset you just saved now appears in the menu straight away, at the top of Recent — no reload needed. Opening the menu also re-checks the list, so presets saved in another tab show up too.

Bulgarian, Indonesian, Malay, and Tamil have been removed from the output language selector. VidNext synchronizes narration to on-screen text by measuring exactly when each word is spoken, and no speech-alignment model exists for these four languages — so videos in them could not be timed correctly and would fail late in generation, after narration had already been produced. Rather than let a video get that far and break, they are no longer offered.

No existing project is affected: none had ever been created in these languages. The selector now offers 27 output languages, and every one of them supports full word-level narration timing.

Pressing Download on a thumbnail variant — in Thumb Lab or on a project's Thumbnails tab — used to just open the image in a new browser tab, leaving you to right-click and save it yourself.

It now starts a real download straight away, the same as Download video does, and the file is named after the variant's concept (for example pointing-at-window-mould.png).

The Quality setting in the thumbnail preset editor — where a preset could choose which image model generated its variants — has been removed.

Every thumbnail run now uses the same image model, the one the service has configured for thumbnails, and automatically falls back to the next one if it is unavailable. Your existing presets keep all their other settings; only the model choice is gone.

Nothing changes about what you pay: thumbnails have always been charged for the model that actually produced the image, and the credit estimate shown before you press Generate already matches it.

In non-English videos, a word-card motion scene (the letter-series style that reveals one word per card) could wrongly discard its correctly generated on-screen words when the narration used a different grammatical form of the same word — common in Ukrainian, Polish, and other inflected languages — and show built-in English sample words instead. The content check now recognizes inflected word forms, keeps real generated content, and if generation still fails, rebuilds the cards from your own scene plan and script (or falls back to a text card with your narration). Template sample text can no longer appear in a finished video in any language, and affected scenes in existing projects fix themselves the next time the visuals step is re-run.

Card timing in non-Latin languages is fixed too: word cards, letter grids, map stops, and stat reveals now sync to the exact moment their word is spoken in Ukrainian and other non-Latin scripts (previously they could all flash at the start of the scene with the last card frozen for the rest of it). When a card's word is a translation or paraphrase of what the narrator says, an extra AI alignment pass matches it to the right spoken moment.

The pricing page now answers the most common questions right below the packages, grouped into five sections: Product & Quality, Credits & Usage, Billing & Payments, Commercial Rights & Watermarks, and Support & Refunds — covering what the visuals are made from, which formats and languages are supported, using your own script or voice, what a finished minute costs in credits, and how to reach support.

The short version of what it covers: credits are spent on AI generation (script, voiceover, visuals) — timeline edits, caption tweaks, and footage swaps are free, with only the export of an updated video using a small amount of credits for rendering; credits bought during the beta never expire; while the beta lasts VidNext runs on flexible top-ups, with monthly subscriptions arriving after it ends; you keep full commercial rights to your videos, and exports carry no watermark. The section is available in both English and Ukrainian.

Some videos generated since August 10 could open in the editor with a few scenes missing from the timeline — most visibly the very first scene — even though the finished video itself was complete and correct. This affected scenes whose footage was sourced from the stock library during generation.

The gap is fixed, and existing projects heal automatically: the next time you open an affected project in the editor, the missing scene blocks are restored in their exact positions with their original footage. Your edits are preserved — the repair only fills the empty gaps, and it never re-adds a scene you deliberately deleted.

Motion-design templates ship with English demo content so they look complete in the design library — and in rare cases a fragment of that demo text could end up in a finished video when the AI intentionally left an optional field empty (for example, a stat card about a topic with no year-over-year comparison could still show a sample "12.5% vs last year" badge). Optional decorative fields now stay empty when the plan leaves them empty: the badge, attribution line, or label simply doesn't render. Demo content is used for library previews only. This applies to every output language.

Videos generated in Japanese, Chinese, or Korean previously failed during script generation — the length check counted these scripts as a handful of words because they don't separate words with spaces, so every attempt was rejected as "too short". Length measurement is now script-aware end to end: scripts in these languages generate at the correct length for the requested duration, scene pacing is sized in characters instead of words, and the narration voice's speaking pace is calibrated on text in the video's actual language.

Captions got the same treatment: Japanese, Chinese, and Korean subtitles now render as natural phrase chunks without spaces between characters, break at the right punctuation, keep to comfortable line lengths, and use proper CJK fonts (including correct Japanese glyph forms instead of Chinese variants). Titles and text animations size their layouts correctly for full-width characters.

Voice previews now speak your selected content language. When you pick a language on the create page, the voice + music sample and the voice auditions in the picker are narrated in that language for all 32 supported languages — no more English preview for a Japanese voice. The English preview script also got a small refresh.

The VidNext blog has moved to the main domain: it's now at vidnext.ai/blog, and the Blog links across the site point there. Existing links to blog posts on the docs domain keep working — they show the same content — but the main-domain address is the one to share and bookmark.

Music video mode has no voiceover — the uploaded song is the whole soundtrack, and the prompt only directs the visuals. Until now, pasting a full narration script there produced a silent beat-cut montage: technically a valid music video, but not what the script's author wanted.

VidNext now checks for this before the video starts. If you're in Music video mode and your text reads like a script meant to be spoken — whether it's pasted into the prompt or attached as a custom script — a dialog explains what the mode will actually do and shows the first line that would never be voiced, alongside your uploaded track.

From there, one click on "Make a narrated video instead" converts the setup for you: your text becomes the narration script, your uploaded track is kept as background music, and the form guides you to pick a voice before generating. Prefer the montage? "Keep Music video mode" continues exactly as configured — the warning never blocks you.

A live hint under the prompt field also appears the moment you paste long text in Music video mode, so you don't have to reach the Create button to find out.

Videos narrated with Cartesia voices now generate their voiceover per act instead of scene by scene. Narration flows smoothly across scene changes with no clipped joins, and every word gets a precise timestamp — so captions, karaoke-style highlights, and word-synced text animations line up with the voice exactly, the same way they already do for ElevenLabs and Fish Audio voices. Burned-in subtitles on Shorts now work with Cartesia voices too.

Redoing the narration step is also cheaper for these projects: acts that didn't change are reused instead of being generated (and billed) again.

In the editor, the narration voice picker now has a Cartesia tab — browse the catalog (pre-filtered to your project's language), preview voices, use your cloned voices, and regenerate any narration with them. And regenerated narrations now carry their own word timings, so "Generate from narrations" in the captions editor stays in sync with the new audio instead of the original take.

Voiceovers no longer stumble over dotted abbreviations. Text like "U.S.", "F.B.I.", or "e.g." used to be read with a long pause after every period — the voice treated each dot as the end of a sentence. Narration is now prepared for speech the same way numbers already are: the voice reads the letters smoothly ("U.S." as "U S", "F.B.I." as "F B I") or expands small abbreviations the way a narrator would ("e.g." as "for example"), in your video's language — while captions, typewriter, and all on-screen text keep the original spelling exactly as written.

Also fixed in the same pass: money amounts with a magnitude word — like "$220.2 million" — could occasionally be voiced with the magnitude doubled ("…million dollars million"). Amounts now read exactly once, the way a narrator would say them.

This applies to every voice and every video type — landscape videos, Shorts, and narrations regenerated in the editor.

On the dashboard, a project's status now sits in the top-left corner of its thumbnail — the first thing you see — and the "how long ago" timestamp moves to the top-right. Projects paused for script approval, or still waiting in the queue, now say so on the card instead of showing a raw state name.

Also: the pricing page drops its film-footage backdrop for the same drifting studio lighting as the home page, in both the light and dark themes.

The studio backdrop's ambient light has been recomposed: cool brand-blue now anchors the top-left corner behind the navigation, while the warm tungsten glow settles into the bottom-right. Same film-style atmosphere, better balance across the page.

Also: when a video is assigned to a Planner channel, the project page's status bar now shows the channel name — click it to jump straight to that channel's dashboard.

The VidNext website (homepage, pricing, about and careers pages) now offers a light theme. A theme switcher in the navigation lets you pick System, Light, or Dark — by default the site follows your device's appearance setting, and your choice is remembered across visits.

The light theme is a warm paper look built around the same film-set identity as the dark one. Product demos, example videos and screenshots keep their true dark interface — the VidNext app itself keeps its cinematic dark look.

VidNext now sits on a warmer, film-style backdrop: the page background has a warm graphite tone with soft drifting light — warm tungsten pools mixed with cool blue ones — behind the dashboard and creation pages. The marketing homepage and the docs site share the same warmer look, and the homepage's animated product demos now show the interface's new colors.

The dashboard header is more compact too: the four project counters (total, completed, in progress, failed) now live as small chips right under the welcome title, and they still work as one-click project filters.

Also: the Ukrainian community Telegram link on the homepage and in the footer now appears for visitors from Ukraine, and the footer now links to VidNext on X.

The official VidNext documentation is now live at docs.vidnext.ai — guides, billing details, changelog, and blog in one place.

Clip selection got noticeably better at staying on-subject. Candidate footage is now scored more strictly against your video's actual topic, so scenes are less likely to pick up look-alike clips from unrelated stories — especially for videos about specific people, events, or franchises.

The YouTube metadata generator now researches competing videos before writing your titles — so title options, tags, and SEO keywords are grounded in what's actually ranking for your topic, not just the script.

Also in this update:

• Ukraine pricing: purchases from Ukraine now get an automatic discount on every top-up, not just the first — applied automatically at checkout.
• Narration fixes: numbers and currency amounts are now read out naturally by narration voices.
• Language selector: fixed cases where the interface and output language selectors disagreed.
• Editor stability fixes.