VEO 3.1

io.github.u2n4Published by GitHub account u2n4v1.1.2Updated Oct 24, 2026

4K video generation via Google VEO 3.1: text-to-video, image-to-video, video extension

VerifiedSTDIODesktop onlyMedia & DesignAI & ML

Overview

AI-generated overview

Local MCP server that lets an assistant generate 4K AI videos with Google VEO 3.1: text-to-video, image-to-video, frame interpolation and video extension.

What you can do
Generate videos from text prompts, with resolution, aspect ratio, duration, negative prompts, reference images, seed and batch options, and animate a reference image with a prompt or interpolate between first and last frames. Extend an existing VEO-generated video by about 7 seconds. Check, list and monitor asynchronous generation jobs, since generation starts in the background and returns a job ID. Inspect API key rotation status, VEO pricing per second and output statistics.
When to use it
Useful when you want an assistant to produce short AI video clips or animate still images through Google VEO 3.1, and to manage the resulting jobs and files from the same conversation.
Requirements
Runs locally as a Python package (pip or uvx). Needs a Gemini API key in GEMINI_API_KEY; an optional GEMINI_API_KEY_BACKUP is used for rotation on rate limits. VIDEO_OUTPUT_DIR sets where videos are saved (default ~/veo-videos). Network access to Google's API is required.
Before you install
Generation calls Google's paid VEO 3.1 API, so usage incurs costs; a pricing tool is provided. The Gemini API key is passed as an environment variable and grants access to the account's quota. Videos are written to a local output directory, and the server deletes stuck processing files before each generation.

Installation

In SourceWeft

  1. Open VEO 3.1 in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

VEO 3.1 MCP Server

Professional-grade MCP server for Google VEO 3.1 video generation with async job pattern, API key rotation, and multi-fallback download.

Quick Install

bash
pip install veo-mcp-server# oruvx veo-mcp-server --help

Claude Desktop (claude_desktop_config.json):

json
{  "mcpServers": {    "veo": {      "command": "uvx",      "args": ["veo-mcp-server"],      "env": {        "GEMINI_API_KEY": "your_key_here"      }    }  }}
VariableDescriptionDefault
GEMINI_API_KEYGemini API key — required for all generation--
GEMINI_API_KEY_BACKUPBackup key for automatic rotation on 429--
VIDEO_OUTPUT_DIRDirectory where generated videos are saved~/veo-videos

What It Does

Generates AI videos using Google's VEO 3.1 model. Supports text-to-video, image-to-video, video extension, and frame interpolation. Uses an async job pattern where generation starts immediately in the background and returns a job ID for polling.

Tools (9)

ToolDescription
veo_generate_videoGenerate video from a text prompt. Supports resolution (720p/1080p/4K), aspect ratio (16:9/9:16), duration (4/6/8s), negative prompts, reference images, seed control, and batch generation (1-4 videos).
veo_image_to_videoAnimate a reference image with a text prompt. Supports resolution, aspect ratio, reference images, and seed control.
veo_interpolate_videoCreate a smooth video transition between a first frame and a last frame image.
veo_extend_videoExtend an existing VEO-generated video by ~7 seconds. 720p only, max 148 seconds total. Source must be from a previous VEO generation (tracks provenance server-side).
veo_check_jobCheck the status of an async video generation job. Call repeatedly until status is 'completed' or 'failed'.
veo_list_jobsList all video generation jobs and their current status.
veo_api_statusShow current API key rotation status — keys configured, active key, keys remaining.
veo_pricing_infoShow VEO 3.1 pricing per second of generated video for both standard and fast models at all resolutions.
veo_show_output_statsDisplay statistics about generated videos and active/recent jobs.

Configuration

VariableDescriptionDefault
GEMINI_API_KEYPrimary Gemini API key (required)--
GEMINI_API_KEY_BACKUPBackup key for automatic rotation on 429--
VIDEO_OUTPUT_DIRDirectory where generated videos are saved~/veo-videos

Architecture

  • Async Job Pattern: All generation tools return a job ID instantly. Background threads handle the long-running Google API calls, so each MCP call returns in under 5 seconds.
  • API Key Rotation: On 429/RESOURCE_EXHAUSTED, automatically switches to the backup key and retries.
  • 3-Layer Download Fallback: SDK download -> URI fallback -> Wait-for-ACTIVE state. Handles Google's Files API edge cases.
  • Preventive Cleanup: Deletes stuck PROCESSING files before each generation to prevent cascading 500 errors.

Models

TierModel IDUse Case
Standardveo-3.1-generate-previewHigher quality output
Fastveo-3.1-fast-generate-previewQuicker generation

License

MIT

Source: servers/veo/README.md at commit 1b8fc5a

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.1.2LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts