Guide · 12 min read · Nov 13, 2026

Chat GPT video generator: can ChatGPT make videos after Sora?

No. OpenAI's video product Sora has been unavailable since April 26, 2026, and according to news reports ChatGPT stopped generating video from text prompts at the same time. A chat gpt video generator today is a two-step workflow: ChatGPT writes the script and prompt, and a dedicated video model renders the clip, starting at 20 credits for 5 seconds.

Desk with laptop, notebook of pencil storyboard sketches, red pencil and coffee mug

If you searched for a chat gpt video generator, you probably want ChatGPT to turn a prompt into a finished clip. This guide covers what changed in 2026, what ChatGPT can still do for a video project, and how to get from a ChatGPT prompt to a real clip on a dedicated video model, such as the ones in Text to Video.

Every date and quote below was checked on November 13, 2026, against OpenAI's own Sora pages and reporting from CNN, Variety, SiliconANGLE, TechCrunch and NBC News. When a detail comes only from news coverage or an independent source, we say so.

Can ChatGPT make videos in 2026?

No, ChatGPT can't make videos as of November 2026, based on OpenAI's Sora pages and news reports. OpenAI's Sora 2 page now carries the banner "As of April 26, 2026, the Sora product is no longer available," and the original Sora page says the same. According to news reports, the shutdown reached ChatGPT too: Variety reported that "ChatGPT also will no longer generate video based on text prompts."

SiliconANGLE reported that ChatGPT "will no longer support video functions either," and that Sam Altman told staff the company would wind down all products that use its video models. We couldn't find an official OpenAI help article confirming the ChatGPT side, so treat that part as reported, not officially stated.

What happened to Sora? Key dates

Sora went from public launch to shutdown in about 16 months, from December 2024 to April 2026. The table shows each step and where the date comes from.

Sora timeline

DateWhat happenedSource type
December 2024First Sora released publicly to ChatGPT Plus and Pro users in the US and CanadaIndependent (Wikipedia)
September 30, 2025Sora 2 released as OpenAI's "flagship video and audio generation model," used in a new Sora appOfficial (OpenAI)
March 24, 2026OpenAI announces it is shutting down the Sora appNews (CNN)
April 26, 2026Sora product no longer availableOfficial (OpenAI Sora pages)
September 24, 2026Sora API discontinuedIndependent (Wikipedia, Opus Clip); not confirmed by an official source

Sources: OpenAI Sora and Sora 2 pages, CNN, Wikipedia, Opus Clip. Checked November 13, 2026.

CNN reported the announcement on March 24, 2026, quoting an OpenAI spokesperson: "As we focus and compute demand grows, the Sora research team continues to focus on world simulation research to advance robotics." TechCrunch noted the app had launched only six months earlier as "a TikTok-like social app."

According to NBC News, Disney said it respects "OpenAI's decision to exit the video generation business," and Sora's head had earlier limited video generation because of a limited supply of chips. No successor has been announced: Opus Clip notes there's no Sora 3, and TechCrunch wrote that OpenAI appeared to be "winding down pretty much everything it's doing with video." Sora isn't available on Klakix either; our Sora model page covers its status.

What ChatGPT can still do for your video

ChatGPT is still useful for everything around the footage: the idea, the script, the shot list and the prompt you paste into a video model. OpenAI's own Sora page now invites readers to "explore image creation in ChatGPT," so stills are still on the table too.

  • Scripts: ask for a 30-second script with a hook in the first line, then read it out with an AI voiceover. Our video script template shows the structure.
  • Shot lists: have ChatGPT split the script into 3 to 6 single shots, one action each, so every shot becomes one generation.
  • Video prompts: turn each shot into a short prompt with subject, action, camera move, light and sound.
  • Images: make a still to use as a first frame. On Klakix, GPT Image 2.5 in the AI image generator costs 7 to 25 credits per image.
  • Captions and hooks: titles and on-screen lines you add later in an editor, since video models often garble text.

How to use a ChatGPT video generator workflow now

The working chat gpt video generator setup in 2026 has two parts: ChatGPT writes, a video model renders. Klakix is not affiliated with OpenAI; it's one place to run the second step, with several video models from Google, ByteDance, Kuaishou, MiniMax and Alibaba and pay-as-you-go credits in USD.

Four-step flow diagram: brief in ChatGPT, one-shot prompt, pick video model, final clip
The two-tool workflow: ChatGPT writes, a video model renders.
  1. Write the brief in ChatGPT: Tell ChatGPT the topic, the audience, the platform and the length, for example a 15-second vertical clip for Instagram Reels.
  2. Ask for one-shot prompts: Ask ChatGPT to turn the idea into single-shot prompts of under 60 words, each with subject, one action, camera move, lighting and sound.
  3. Open Text to Video: Go to Text to Video on Klakix and paste the first prompt into the prompt box.
  4. Pick model, length and ratio: Choose a model, then a length and aspect ratio, such as Google Veo 3.1 Lite at 8 seconds in 9:16.
  5. Generate and review: Generate the clip, watch it, and tweak one detail in the prompt if the motion or light is off.
  6. Assemble in the Video Editor: Join your clips in the Video Editor and add captions, music or a voiceover there before exporting.

Klakix's generation tools make one clip at a time; joining clips and adding text, music or voice happens in the Video Editor, where editing and MP4 export are free. If a generation fails, its credits are refunded automatically.

ChatGPT video generator prompts you can copy

The best ChatGPT prompt for video asks for one shot, not a whole story, because each generation is a single continuous clip. Copy this into ChatGPT:

Template: "Write one prompt for an AI video model, under 60 words, describing a single 8-second shot. Include the subject, one clear action, the camera move, the lighting, the setting and the sounds. No on-screen text and no real people. Topic: [your product or idea]. Format: 9:16 vertical."

Typical output: "Extreme close-up of black coffee pouring into a white ceramic mug on a wooden counter. Steam curls up in low morning window light. The camera pushes in slowly. Soft cafe murmur and the sound of pouring. Vertical 9:16." This works because it gives the model one action, one camera move and a sound cue, and it leaves out text that video models tend to misspell.

For an explainer with narration and exact numbers, ask ChatGPT for a topic outline instead and use Motif, which builds a 30 to 90 second video with voiceover, captions and motion graphics from a topic for 139 to 217 credits. For product ads with a presenter, UGC Ads takes a 320-character script.

Extreme macro of tea poured from a teapot into a tulip-shaped glass. Still from a video made with Google Veo 3.1 Lite.
Google Veo 3.1 LiteExtreme macro of tea poured from a teapot into a tulip-shaped glass. Steam rises, backlit by low morning sun, with a blurred waterside tea garden behind. Slow motion, shallow depth of field, film grain. The pour and a spoon clinking against the glass.
A ferry crossing the Bosphorus with seagulls overhead, golden hour light. Still from a video made with MiniMax H3 Max Turbo.
MiniMax H3 Max TurboA ferry crossing the Bosphorus with seagulls overhead, golden hour light

Each sample was generated in one take on this site with the model shown on it; nothing was edited.

Can ChatGPT turn an image into a video?

ChatGPT can make the image but not the motion, so the chat gpt image to video route needs a second tool. Save the image, upload it to Image to Video, and describe how it should move, such as "steam rises and the camera circles slowly."

One limit catches people: the Seedance models don't accept photos with human faces. For portraits, MiniMax H3 Max (40 credits for 5 seconds at 768p) is the model built to keep faces and details intact. Only use photos of people who agreed to it; Klakix asks for consent and doesn't allow photos of celebrities or politicians. Our guide on how to animate a photo covers motion prompts in detail.

Is there a free ChatGPT video generator without a watermark?

No free ChatGPT video generator exists in November 2026, because OpenAI's video product, Sora, is no longer available. Even when Sora was live, its videos carried a visible, moving watermark, according to Wikipedia. Skip any site offering a free ChatGPT video generator download or APK: OpenAI has no video app to download, and cracked apps can carry malware.

Many "ChatGPT video" tools are script-to-video editors rather than generative models. Pictory says it has its own GPT in the GPT Store that writes a script and makes a video, and Fliki describes its flow as "ChatGPT writes the script. Fliki turns it into the finished video," in 9:16 or 16:9 MP4. Those tools assemble footage, voice and captions; a generative model creates new footage from your prompt. Both are valid, they just do different jobs.

On Klakix, no visible watermark is added and commercial use is allowed under the terms of service. Video isn't free, though: the 20 credits you get after verifying your email work only for voiceovers, and video unlocks after your first credit purchase.

How much does an AI video cost without Sora?

A 5-second AI clip starts at 20 credits on Klakix, which is about $1.33 on the $5 Mini pack (75 credits, so about 6.7 cents per credit, our math). An 8-second Veo 3.1 Lite clip at 40 credits comes to about $2.67 on the same pack.

Text to Video models at default settings

ModelLengthsAspect ratiosSoundDefault costAbout on Standard pack
MiniMax H3 Max Turbo5, 10, 15 sec16:9, 9:16, 1:1, 4:3, 3:4, 21:9Yes20 credits (768p, 5 sec)$1.02
Google Veo 3.1 Lite4, 6, 8 sec16:9, 9:16Optional, dialogue and ambient40 credits (720p, 8 sec)$2.04
Seedance 2.0 Mini4 to 15 sec16:9, 9:16, 1:1, 4:3, 3:4, 21:9Optional48 credits (720p, 5 sec)$2.44
Kling 4.03 to 30 sec16:9, 9:16, 1:1, 21:9Always, stereo79 credits (720p, 5 sec)$4.02
Wan 3.05 to 30 sec16:9, 9:16, 1:1, 4:3, 3:4Optional100 credits (720p, 10 sec)$5.09
Seedance 2.54 to 30 sec16:9, 9:16, 1:1, 4:3, 3:4, 21:9Optional147 credits (720p, 5 sec)$7.48

Source: Klakix catalog. Dollar column is our math on the Standard pack ($14 for 275 credits, plus tax). Checked November 13, 2026; may change.

Longer clips and Full HD cost more than these defaults. Kling 4.0 and the higher-priced models need more than the 75 credits in a Mini pack, so a bigger pack or the Starter plan ($9 a month for 200 credits) fits better; see pricing for every option. Google's video models on Klakix are covered on the Veo model page.

Sources

The facts in this post were checked against the pages below on Nov 13, 2026. Prices and limits may change.

  1. OpenAI: Sora 2 (official source)
  2. OpenAI: Sora (official source)
  3. Variety: OpenAI shutting down Sora (news)
  4. SiliconANGLE: OpenAI pulls the plug on Sora (news)
  5. CNN: OpenAI Sora video app shutting down (news)
  6. TechCrunch: Sora is shutting down (news)
  7. TechCrunch: Sora's shutdown and AI video (news)
  8. NBC News: OpenAI shuttering Sora (news)
  9. Opus Clip: Sora 2 alternatives after shutdown (independent source)
  10. Pictory: ChatGPT video creator (independent source)
  11. Fliki: ChatGPT video generator (independent source)

This post was drafted with AI assistance. Facts were checked against the sources above, and model, credit and price details were checked automatically against the site's current price list.

Frequently asked questions

What is the ChatGPT video generator called?

OpenAI's video generator was called Sora, with Sora 2 released in September 2025. OpenAI's own pages say the Sora product has been unavailable since April 26, 2026. There's no new name for a video feature in ChatGPT, because no successor has been announced.

Is the chat gpt video generator free?

There's no ChatGPT video generator to use, free or paid, as of November 2026. ChatGPT can still help with scripts, prompts and images. To render the clip, you need a separate video model, and most charge per generation.

How do I access the ChatGPT video generator?

You can't, because OpenAI shut down Sora and news reports say ChatGPT stopped generating video at the same time. The practical route is to have ChatGPT write a one-shot prompt, then paste it into a text-to-video tool that runs current models.

Is there a ChatGPT video generator app?

No. The Sora app was OpenAI's video app and it closed on April 26, 2026. Apps or download sites claiming to be a ChatGPT video generator are third-party products, and cracked versions are best avoided.

ChatGPT vs Gemini: which one can make videos?

ChatGPT can't generate video anymore, according to news reports. We haven't verified what the Gemini app itself offers for video or at what limits. Google's video models Veo 3.1 Lite and Gemini Omni Flash are available on Klakix if you want to try Google's video output.

Can ChatGPT make a video from an image?

ChatGPT can create the still image, but it doesn't animate it. Upload that image to an image-to-video model and describe the motion you want. Some models reject photos with human faces, so check the model notes before uploading a portrait.

Is Sora coming back as Sora 3?

OpenAI hasn't announced a successor, according to Opus Clip. Independent sources also report that the Sora API ended on September 24, 2026, though that date isn't confirmed on an official OpenAI page. For now, plan your video work around other models.

How much does it cost to make an AI video without Sora?

On Klakix, a 5-second clip starts at 20 credits with MiniMax H3 Max Turbo, about $1.33 on the $5 Mini pack. Premium models cost more, such as 147 credits for Seedance 2.5 at its default 5 seconds in 720p, and longer or Full HD clips cost more still. Prices were checked November 13, 2026 and may change.

Try what you just read

Signing up is free. Video and image tools unlock after your first credit pack, and credits are refunded if a generation fails.

Open Text to Video