Google Vids MCP Server: Gemini Omni Video and Voiced Avatars

Google Vids MCP Server

October 16, 2026 (November 4, 2026)

Endpoint and authorization
Setup
Use it in the Claude app or ChatGPT
Tools
Long-running work
Example prompts
Limits and notes

The useapi.net Google Vids MCP server lets an AI assistant such as the Claude app, ChatGPT, Claude Code, Cursor or VS Code Copilot make videos with Google Vids on your own Google accounts. It is a hosted Model Context Protocol server with 18 tools: generate Gemini Omni 1.1 Flash clips with sound and speech from text, a start image or up to 3 references, extend, upscale and edit them, make images, upload your own pictures, create voiced avatars from a description, a photo or another avatar, make an avatar read a script as a talking-head video, and turn a script into speech. Each tool calls the Google Vids API, so the same rules and limits apply. Nothing needs to be installed or hosted.

Is there a Google Vids MCP server? Yes. useapi.net, a third-party service not affiliated with Google, hosts one at https://api.useapi.net/v1/google-vids/mcp with 18 tools for Gemini Omni 1.1 Flash video, images, talking avatars and speech, on your own Google AI Pro or Ultra plan, nothing to install.

Prefer REST? Tutorial: How to Make AI Influencer Videos with Avatars via the Google Vids API · Code: github.com/useapi/google-vids-api

See all useapi.net MCP servers on Model Matrix & MCP.

Endpoint and authorization

https://api.useapi.net/v1/google-vids/mcp

Transport: Streamable HTTP.
Listed in the official MCP Registry as net.useapi/google-vids, so MCP clients and directories that read the registry can find it by that name.
Authorization: your useapi.net API token in the Authorization: Bearer … header, see Setup useapi.net. The Claude app and ChatGPT sign in with OAuth instead, see Use it in the Claude app or ChatGPT.
You need at least one Google account added with Setup Google Vids. Adding and removing accounts is not available over MCP: use POST /accounts or the setup page.

Setup

Replace YOUR_API_TOKEN with your useapi.net API token.

Claude Code

claude mcp add --transport http useapi-google-vids https://api.useapi.net/v1/google-vids/mcp --header "Authorization: Bearer YOUR_API_TOKEN"

Cursor

In ~/.cursor/mcp.json:

{
  "mcpServers": {
    "useapi-google-vids": {
      "url": "https://api.useapi.net/v1/google-vids/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_TOKEN"
      }
    }
  }
}

VS Code

In .vscode/mcp.json:

{
  "servers": {
    "useapi-google-vids": {
      "type": "http",
      "url": "https://api.useapi.net/v1/google-vids/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_TOKEN"
      }
    }
  }
}

Windsurf and other clients

Any client that supports remote MCP servers over HTTP with custom headers works. The usual JSON shape:

{
  "mcpServers": {
    "useapi-google-vids": {
      "serverUrl": "https://api.useapi.net/v1/google-vids/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_TOKEN"
      }
    }
  }
}

Some clients name the address field url instead of serverUrl.

Test it with curl

curl -s https://api.useapi.net/v1/google-vids/mcp \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"curl","version":"1"}}}'

curl -s https://api.useapi.net/v1/google-vids/mcp \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/list"}'

The first call answers with the server’s name useapi-google-vids and its instructions for the assistant, the second with the 18 tools.

Use it in the Claude app or ChatGPT

The Claude app (claude.ai and Claude Desktop) and ChatGPT can add this server as a custom connector. They sign in with OAuth: you give the app the URL, it opens a useapi.net page, and you paste your useapi.net API token there once. Both work on their Free plans. The API-token setups above for Claude Code, Cursor, VS Code and other clients keep working as before.

https://api.useapi.net/v1/google-vids/mcp

Claude app

Open Settings → Connectors and click Add custom connector.
Enter a name, for example useapi Google Vids, and the URL above, then click Add.
Click Connect. Claude opens the useapi.net sign-in page.
Paste your useapi.net API token and click Allow.

Claude asks before each tool call unless you choose Always allow. The Free plan allows one custom connector.

ChatGPT

Open Plugins, click Add and choose Add custom MCP server.
Enter a name and the URL above, and set Authentication to OAuth.
Tick ChatGPT’s “I understand and want to continue” box, then click Create as a plugin.
Click Connect. ChatGPT opens the useapi.net sign-in page.
Paste your useapi.net API token and click Allow.

ChatGPT asks you to confirm each tool call that creates or changes something. The Free plan works without developer mode (as of October 25, 2026).

Your API token is in your welcome email, see Setup useapi.net. The useapi.net page links there too.
Each connection is for one MCP server. Connect each server you want to use on its own.
The app gets an access token that lasts 1 hour and renews it on its own. A connection you don’t use for 90 days lapses, and you connect again.
Resetting your API token disconnects every app connected with it.
An image also comes back as an image the app can see. Videos and image files come back as download links that work in your browser without sign-in for 1 hour. Google keeps the files only for a while, so download them soon.

Tools

Clips, images, uploads and avatars are named by the ids the tools return, and each id carries its account, so a follow-up runs on the account that holds its inputs. Tools that start a job take an optional email to pick the account.

Accounts

Tool	What it does	Main parameters
`vids_list_accounts`	Lists your accounts with health, `maxJobs` and the plan’s monthly limits.
`vids_get_account`	Checks one account live: signed in or not, and what is left of this month’s video seconds, images, AI avatar videos (`avatarVideo`) and Gemini uses (`speech`).	`email`
`vids_update_account`	Sets `maxJobs`, the jobs an account runs at once.	`email`, `maxJobs`

Videos and images

Tool	What it does	Main parameters
`vids_create_video`	Generates a 3 to 10 second clip with sound and speech from text, from a start image, or from up to 3 references (images and avatars, named in the prompt as `@referenceImage_1` or `@avatar_1`).	`prompt`, `duration`, `aspectRatio`, `resolution`, `startImage`, `referenceImage_1`..`3`, `avatar_1`..`3`, `email`
`vids_extend_video`	Adds 3 to 10 seconds to a clip and returns the whole clip, up to about 41 seconds.	`mediaId`, `prompt`, `duration`
`vids_upscale_video`	Turns a `720p` clip of up to 20 seconds into `1080p`.	`mediaId`
`vids_edit_video`	Changes a clip by prompt: colours, objects, style.	`video`, `prompt`, `duration`
`vids_create_image`	Generates one JPEG in one of 7 styles, returned as an image the assistant can see.	`prompt`, `aspectRatio`, `style`, `email`
`vids_create_avatar_video`	Makes an avatar read a script in its own voice: a 1280×720 talking-head video as long as the speech, up to 30 seconds (script up to 400 characters, audio tags like `[whisper]` allowed). Uses 1 AI avatar video of the account, also when Google’s safety check refuses the finished clip. Takes 1 to 2.5 minutes, so it usually returns the `jobid`.	`avatar`, `script`
`vids_create_speech`	Turns a script of up to 2,500 characters into a WAV in one of the 30 voices, to hear a voice before a talking avatar video or to use the audio elsewhere. Uses 1 Gemini use of the account.	`script`, `voice`, `email`
`vids_get_job`	Reads a video, image, speech or talking avatar video job, and with `wait: true` waits for it to finish.	`jobid`, `wait`
`vids_list_jobs`	Lists the jobs running now and each account’s running count against `maxJobs`.

The parameters match POST /videos, POST /videos/extend, POST /videos/upscale, POST /videos/edit, POST /videos/avatar, POST /images and POST /speech.

Files, avatars and voices

Tool	What it does	Main parameters
`vids_upload_asset`	Uploads your own JPEG, PNG or WebP image from a public https URL of up to 50 MB, or sent as `contentBase64`, and returns an `assetId` for `startImage` or `referenceImage_N`, or for a person’s photo or a logo in `vids_create_avatar`. Upload with the `email` of the account you will use it on.	`url` or `contentBase64`, `mimeType`, `email`
`vids_get_media`	Checks a video, image or speech `mediaId` and returns its download URL. An image also comes back as an image the assistant can see.	`mediaId`
`vids_list_voices`	Lists the 30 Gemini voices with their style and preview audio.
`vids_list_avatars`	Lists the avatars saved on your accounts, including ones made in the Google Vids web app.	`email`
`vids_create_avatar`	Makes a new voiced avatar from exactly one start: a description (`appearance`), a person’s photo from `vids_upload_asset` or an image from `vids_create_image` (`image`), or an existing avatar (`avatar`). `refine` and `logo` then change the picture. Each picture Google makes counts 1 image where the plan has an image allowance, and costs nothing on a plan without one. Google refuses photos of prominent people. The new picture comes back as an image.	`name`, `voice`, `appearance`, `outfit`, `shot`, `expression`, `details`, `image`, `avatar`, `refine`, `logo`, `email`
`vids_delete_avatar`	Deletes an avatar.	`avatarId`

The full input schema of every tool comes from tools/list.

Long-running work

MCP clients usually give up on a tool call after about a minute, so a tool waits up to about 50 seconds:

An image usually finishes in about 10 seconds, and the tool returns it.
A 720p clip often finishes in that time. A 1080p clip, an extend, an upscale or an edit takes longer. The tool then returns the jobid, and vids_get_job with wait: true picks it up. Ask the assistant to keep checking until it is done.
vids_create_avatar answers in about 10 to 60 seconds, the longest when a photo or description is also refined.
vids_create_speech usually finishes in a few seconds.
vids_create_avatar_video takes 1 to 2.5 minutes, so it returns the jobid: vids_get_job with wait: true picks it up.

Clients that ask for progress notifications get one about every 10 seconds while a tool waits.

Clips come back as a mediaId with a download URL. Fetch it with the same Authorization header, and download soon, since Google keeps the files only for a while, see How long files last.

Example prompts

“Make a 10-second 1080p clip of a fox running through a snowy forest at dawn.”
“Make a photo of a red bicycle in a Lisbon street, then turn it into a clip where someone rides it away.”
“Create an avatar of a cheerful chef in her fifties with the Neo voice, then make a portrait clip where she says: ‘Today we bake bread.’”
“Upload this photo of me, make it an avatar in a navy blazer with the Holt voice, and have it read this 20-second product intro.”
“Read this script in the Kaci voice so I can hear it before I make the avatar video.”
“Upload this product photo and make a clip of my avatar holding it and talking about it.”
“Extend the last clip by 8 seconds with the camera pulling back, then upscale it to 1080p.”
“How many video seconds are left on my Vids accounts this month?”

Limits and notes

The tools use your Google accounts and the same rules as the REST API, such as maxJobs. Video seconds, images, AI avatar videos and Gemini uses (speech) count against the account’s monthly Vids allowance, see Plans and monthly allowances.
A client may send up to 20 JSON-RPC requests in one batch.
If the client sends an MCP-Protocol-Version header, it must be one of the supported versions: 2025-06-18, 2025-03-26 or 2024-11-05.
Tool calls show in your usage stats (bot=google-vids) through the REST call each tool makes, the same rows a direct API request writes.