Can ChatGPT edit videos? What ChatGPT and Claude can do in October 2026

What ChatGPT can do with a video file

As of October 2026, ChatGPT can accept video files as attachments. OpenAI's help center says it takes them through supported upload methods, including on the Free plan, and that what you can do varies by platform, upload method and account (OpenAI, image inputs FAQ).

The same page gives a warning. ChatGPT may use tools to analyze an uploaded video, but the analysis can be incomplete or inaccurate. It may not analyze the whole video, and it may not read the audio correctly. So you can ask what happens in a video, but check the answer against the footage.

Audio has its own rules. OpenAI's upload page says audio uploads need a paid ChatGPT plan. It also says WebM and MP4 files identified as video are not supported by audio uploads (OpenAI, uploading files). If you only want a transcript, take the sound out of the video first. The free video to MP3 tool does that in your browser.

The same upload page lists uses such as summarizing a file, finding specific information in it and rewriting it. Applied to a transcript, that gives you a short summary, a list of moments worth keeping and draft captions. You then make the cuts yourself in an editor.

Is there a ChatGPT video editor in the chat itself?

OpenAI's two upload pages describe asking questions about files. Neither one describes cutting a video, adding captions to it or exporting a new video file. So as of October 2026 there is no documented way to ask ChatGPT alone for a finished, edited video. The chat is not a video editor by itself.

Some people think of Sora here. OpenAI introduced Sora as a model that creates videos from text. Its page now says the Sora product is no longer available as of April 26, 2026 (OpenAI, Sora).

What does change the answer is a connected tool. On August 6, 2026, Adobe announced a plugin for ChatGPT. Adobe says it can turn a long video into a highlight reel and adapt it for YouTube Shorts, Instagram Reels and other social platforms. Adobe also says that when you want advanced editing, you can continue your work in its apps. The plugin is available on ChatGPT web and in the ChatGPT app.

That is the pattern to look for in ChatGPT video editing. The chat takes your request in plain words. A connected tool does the video work and hands the result back.

Can Claude edit videos?

As of October 2026, Claude's help center lists the files you can upload to a chat. The documents are PDF, DOCX, CSV, TXT, HTML, ODT, RTF, EPUB, JSON and XLSX. The images are JPEG, PNG, GIF and WebP. Video is not on the list (Anthropic, uploading files). The page showed an update date of July 23, 2026.

Claude can connect to outside tools, though. As of October 2026, its help center says custom connectors that use remote MCP are available on the web and in Claude Desktop for the Free, Pro, Max, Team and Enterprise plans. Free users can add one custom connector (Anthropic, custom connectors).

Claude Code is Anthropic's coding tool, and it connects to outside tools the same way. Its documentation shows one command for adding a remote HTTP server: claude mcp add --transport http <name> <url> (Anthropic, Claude Code and MCP).

Adobe also made a plugin for Claude. Its post of September 24, 2026 says the plugin gives access to more than 80 tools, and it lists editing and assembling videos among the things the plugin does. So the answer for Claude video editing matches the answer for ChatGPT. On its own, Claude's upload page covers documents and images. With a connected editor it can direct a video edit.

Why a chatbot needs an editor behind it

Editing a video file is several jobs. Software has to decode the video, cut it at the right frames, draw captions on top, crop it to a new shape and write a new file. A chat window sends and reads text. To do that work, it has to call software that handles video.

That is what a connector is for. The chatbot decides what should happen and sends a request. The editor does the video work and sends back a result. The chatbot then tells you what it did.

This also explains why the result depends on the editor. A chatbot can ask for a cut, a caption style or a crop. Whether the cut lands on the right frame and the captions match the audio is decided by the software behind the chat.

A text answer is still useful before you edit. A plan, a list of moments and a set of titles all save time, but none of them is a video.

How an assistant connected to an editor works

The standard that connects them is MCP, the Model Context Protocol. Anthropic describes it as an open source standard for AI-tool integrations. An editor that offers an MCP server lets any assistant that supports MCP use its tools.

In practice it goes like this:

1. You add the editor's MCP address to your assistant once.
2. You ask for an edit in the chat, in plain words.
3. The assistant calls the editor's tools, for example to find the best moments or to export one.
4. The editor does the work, and the assistant tells you when it is done.
5. The assistant gives you a download link, and you check the file.

A tool that spends money should ask first. In GPT-Video's MCP server, the tools that spend credits need an explicit confirm, so the assistant asks you before it spends. If you ask twice for the same thing, you get the work already running and you pay once.

How GPT-Video fits: describe the edit in chat

GPT-Video is an AI video editor that you direct by chat. You upload a video, then describe the change you want to the Editor Assistant, for example "cut the first eight seconds, they are only setup". The assistant applies it to the clip on screen, and every change can be undone.

This way of editing is called vibe editing. The vibe editing page shows how it works, and the 90-second film shows it on screen. If you want the background first, what is vibe editing explains it step by step.

GPT-Video is a vibe editing tool. As in ChatGPT, you say what you want in plain words and the tool does the work. If you start in ChatGPT and want the edit itself, the next step is from ChatGPT to GPT-Video: the same kind of chat, with an editor behind it. GPT-Video is an independent product and is not affiliated with OpenAI.

Some limits are worth knowing. You upload the video file itself, because GPT-Video does not import from a link. It stops at the exported MP4 file, so you post the file yourself.

Using GPT-Video from Claude or ChatGPT

The same product can be reached from an assistant through the GPT-Video MCP server. It comes in two editions. The directory edition has eight tools for finding clips and exporting them as vertical MP4 files. It is the edition that Claude on the web and on the desktop, and ChatGPT, connect to by address.

The full edition has the whole catalogue, including a toolset for editing a finished video and asking the Editor Assistant. It is for Claude Code, Cursor, VS Code, Windsurf and n8n, and it needs an API key.

The directory tools cannot upload a file. The assistant gets a link, and you open it to add the video in your browser. In the full edition, an assistant that has a shell can upload for you.

Your plan with each assistant matters too. OpenAI's help page says custom MCP server connections are available to ChatGPT Business and Enterprise/Edu customers on ChatGPT web, as of October 2026 (OpenAI, custom MCP servers). Check it for your plan.

Finding clips spends credits, and an export uses one of the day's exports. An account can get an API bonus of 50 credits with no card, and a limited number are given each day. After the bonus, MCP and the API need a Creator or Studio plan. API pricing has the details, and Use GPT-Video from an AI agent has a prompt you can copy.

Chatbot, connected editor or both?

Use a chatbot alone for text work. That means summarizing a transcript, listing the moments worth keeping, and writing titles and captions. It needs nothing more than the chat.

Use an editor when you want a finished video file. You can work in the editor's own chat, or connect the editor to the assistant you already use. Both routes end the same way, with an MP4 file you can review.

Whichever you choose, watch every clip before you post it. A connected editor makes the cuts quickly, and a person still decides which ones are worth posting. The pillar guide on AI video editing explains where the machine does well and where it needs help.

If you want to try the editor route, the AI clip generator is the place to start, and pricing lists the plans.