google/veo
MCP ServerGoogle Veo MCP Server - text-to-video and image-to-video via Vertex AI / Gemini API
check: npx https://ai.google.dev/gemini-api/docs/video
Google Veo 3 video generation MCP server. Text-to-video and image-to-video with native dialogue and SFX via the Gemini API or Vertex AI.
MCP Server
- Package:
google-veo3-1-mcp-server(community wrapper around the Gemini Video API) - Transport: stdio
- Command:
npx -y -p google-veo3-1-mcp-server veo3-mcp-server
Authentication
API key (Gemini Developer API) or Application Default Credentials (Vertex AI).
| Variable | Required | Description |
|---|---|---|
GOOGLE_API_KEY | Yes (Gemini path) | Gemini API key from aistudio.google.com |
GOOGLE_APPLICATION_CREDENTIALS | Yes (Vertex path) | Path to a Google Cloud service-account JSON with Vertex AI access |
GOOGLE_CLOUD_PROJECT | Yes (Vertex path) | GCP project ID for Vertex |
GOOGLE_CLOUD_REGION | No | Vertex region (default us-central1) |
Available Tools
| Tool | Description |
|---|---|
generate_video | Text/image → video via Veo 3.1 |
extend_video | Extend an existing generation |
interpolate_frames | Frame interpolation |
get_video_status | Poll a generation job |
Configuration Example
[[mcp.servers]]
name = "veo"
type = "stdio"
command = "npx"
args = ["-y", "google-veo3-1-mcp-server"]
timeout_seconds = 120
env = { GOOGLE_API_KEY = "{{INPUT:GOOGLE_API_KEY}}" }
tools = []Notes: Veo 3 produces video with synchronized speech and ambient audio. Output is paid-tier only — confirm pricing in Vertex / Gemini before running production batches. Long generations are async; poll veo_get_operation until done.