ChatGPT in November 2026: Images 2.6 makes GIFs, uploads take documents up to 512 MB and images up to 20 MB but no video, and the Sora API closed on September 24. ReelSDK renders the MP4 for $0.03 per output minute.

Can ChatGPT make a video?

No, ChatGPT gives you text, code and images, not an MP4, and OpenAI shut Sora's app on April 26, 2026. ReelSDK does the rendering: paste a script from ChatGPT and get an MP4 short for $0.30.

LearnUpdated

At a glance

What ChatGPT can do for a video

As of November 2026.

What ChatGPT can do for a video
Video taskCan ChatGPT do it on its own?What to use instead
Write a script or shot listYesChatGPT can handle the text
Draft caption textYesA renderer or editor puts it into the video
Create a finished video fileNoA video renderer or editor
Write ffmpeg commands or Remotion codeYes, but it doesn't run them into a video for youRun the commands or code elsewhere
Turn photos into an MP4 with musicNoA slideshow maker
Make a GIFYes, with ChatGPT Images 2.6Use it when a GIF is enough
ReelSDK APIRenders captions, slideshows and shorts from one API call$0.03 per output minute; shorts $0.30

Sources: OpenAI: the Sora discontinuation, OpenAI: uploading files to ChatGPT, checked November 2026.

The short answer

ChatGPT doesn't create video files as of November 2026. You can ask it to write a video script, plan the shots, draft captions or write code for a video, but you'll need something outside ChatGPT to turn that work into a finished file.

That's the distinction. Getting a script or a block of code in the chat isn't the same as getting a video you can download and post.

What about animation? ChatGPT Images 2.6, released on November 4, 2026, can make GIFs. If a GIF is what you want, that's an option inside ChatGPT. It's separate from making a video file.

What happened to Sora

OpenAI shut down the Sora app and website on April 26, 2026. The Sora API shut down on September 24, 2026.

You can't use those Sora services now. If you're following instructions that send you to the Sora website or tell you to use its API, those instructions no longer describe an available route.

And a ChatGPT conversation doesn't replace the closed services: asking ChatGPT to generate a video still won't give you a video file. You can keep working on the idea in the chat, but the rendering has to happen elsewhere.

What ChatGPT can do for video

Start with the part you need help with. You can ask for a script, a shot list or caption text, without asking ChatGPT to handle the whole production.

For a script, tell it what the video is about and who's going to watch it. For a shot list, ask it to turn your script into a plan you can work from. If the wording is already settled, ask for captions rather than another rewrite.

For example: "Turn this script into a shot list. Keep the spoken wording as written, and put the caption text beside the relevant shot."

ChatGPT can also write ffmpeg commands and Remotion code. Those are things you run somewhere else, not finished videos attached to its reply. Ask for whichever format you intend to use, rather than leaving the output format open.

There's a limit to keep in mind when bringing material into the chat. ChatGPT's file upload help lists documents up to 512 MB and images up to 20 MB; it doesn't list a video file type. Don't plan your editing workflow around uploading a video and having ChatGPT return an edited copy.

Turning ChatGPT's help into a real video

Once you've got the script or captions you want, move to the tool that will render the video. If you've asked for ffmpeg commands or Remotion code, you'll need to run that output outside ChatGPT.

You can ask ChatGPT to revise the commands or code, too. Keep the request specific: change the caption wording, use the revised script, or update the shot list. The output is still instructions or code until you run it.

ChatGPT's Python environment for data analysis can't make external web requests or API calls, so it isn't a way to reach an online rendering service from inside the chat.

If you want an API-based route, ReelSDK renders the video ChatGPT helps you plan. Captions and slideshows cost $0.03 per output minute; faceless shorts cost $0.30, with the first $0.50 free.

Its MCP server gives AI agents 6 video tools. But that local server runs from Codex, not directly from ChatGPT on the web (which uses remote MCP servers through plugins). Keep that distinction in mind if you're choosing between working in the browser and using Codex.

Photos to video

If your starting point is a set of photos, you can use ChatGPT to plan their order and write the caption text, then take that plan to a slideshow tool.

Keep the request focused on the photos and the video you want to make, for example: "Help me put these photos in order for a slideshow and draft the caption text. Give me a written plan, not code."

For the finished file, ReelSDK's free slideshow maker turns up to 10 photos into an MP4 with music, with no signup.

You don't need to ask ChatGPT for rendering code if that slideshow maker covers what you want. Use the written plan, choose your photos and make the MP4 there.

FAQ

Questions, answered

Can I download a video made by ChatGPT?

Not as of November 2026. ChatGPT can prepare text, commands and code for a video, but it doesn't create the finished video file.

Can ChatGPT edit a video I upload?

ChatGPT's file upload help doesn't list a video file type. It can write editing commands or code for you to run elsewhere, rather than return an edited video from the chat.

Is Sora still available?

No. OpenAI shut down the Sora app and website on April 26, 2026, and the Sora API on September 24, 2026.

Can ChatGPT make animated images?

ChatGPT Images 2.6 can make GIFs. That's an option for a GIF, not a way to export a finished video.

Ready to build?

Ship video from your app today

Your first $0.50 is free, and one key covers every endpoint.

Moonlit lake with a lit cabin under the mountains