Skip to main content
Glama
Sign In
@runapi.ai/gemini-omni-mcp

@runapi.ai/gemini-omni-mcp

by runapi-ai
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Why This Package?

@runapi.ai/gemini-omni-mcp is a focused Model Context Protocol server for the Gemini Omni model line on RunAPI. It gives MCP-compatible assistants direct access to 3 endpoints and 5 model variants without loading the full RunAPI catalog.

Use this per-model server when an agent should stay scoped to Gemini Omni. Use @runapi.ai/mcp when one assistant should discover every RunAPI model line.

Related MCP server: imagen-4-mcp

Install

Add it to Claude Code:

claude mcp add gemini-omni -s user -- npx -y @runapi.ai/gemini-omni-mcp

Use project scope when the server should be shared with a repository:

claude mcp add gemini-omni -s project -- npx -y @runapi.ai/gemini-omni-mcp

Codex, Cursor, Windsurf, VS Code, Roo Code, and other MCP hosts can use the same stdio command:

{
  "mcpServers": {
    "gemini-omni": {
      "command": "npx",
      "args": ["-y", "@runapi.ai/gemini-omni-mcp"]
    }
  }
}

check_pricing works before sign-in. For task creation and status polling, ask your assistant to call the login tool. It opens a browser login and saves credentials to ~/.config/runapi/config.json, the same file used by runapi login. Headless and CI hosts can still set RUNAPI_API_KEY before starting the MCP host.

Ready-made examples are in examples/ for Claude, Cursor, Windsurf, VS Code, and Roo Code.

Tools

Tool

Auth

Purpose

create_audio

Yes

Run a Gemini Omni create audio operation synchronously. Returns the operation result.

create_character

Yes

Run a Gemini Omni create character operation synchronously. Returns the operation result.

text_to_video

Yes

Create a Gemini Omni text to video task and optionally wait for a terminal status. Returns the task id, status, and output URLs.

get_task

Yes

Fetch the current status and latest payload for an existing task.

check_pricing

No

Look up current pricing for a Gemini Omni model and endpoint.

Models

Gemini Omni covers 5 model variants across 3 endpoints. Each tool accepts the models listed for it:

Tool

Models

create_audio

gemini-omni-audio

create_character

gemini-omni-character

text_to_video

gemini-omni-flash-1-1, gemini-omni-flash-preview, gemini-omni-text-to-video

Model availability can change between releases. Use check_pricing or the Gemini Omni model page for the current catalog view.

Agent Prompts

Ask your assistant in natural language; it can inspect pricing, create the task, and return the task id plus output URLs.

Create a task

Run a Gemini Omni text to video task with RunAPI.

The assistant can call check_pricing, then text_to_video, and return the task id, status, and output URLs.

Submit without waiting

Create the task but don't wait for it to finish.

The assistant calls the create tool with wait: false and returns the task id. Check on it later with get_task.

Check pricing before creating

Check current Gemini Omni pricing, then create the task if it matches my request.

The assistant calls check_pricing and can link to the Gemini Omni model page for the canonical catalog entry.

Configuration

The server resolves auth in this order:

  1. RUNAPI_API_KEY environment variable, useful for headless and CI hosts

  2. ~/.config/runapi/config.json, created by the MCP login tool or runapi login

  3. No key, which still allows check_pricing

The config file is normally managed by login. A pre-provisioned headless config can use:

{
  "apiKey": "your_runapi_key"
}

Do not commit real API keys.

Links

Resource

URL

Gemini Omni model page

https://runapi.ai/models/gemini-omni

npm package

@runapi.ai/gemini-omni-mcp

GitHub repository

runapi-ai/gemini-omni-mcp

RunAPI MCP overview

runapi.ai/mcp

RunAPI docs

runapi.ai/docs

License

Licensed under the Apache License, Version 2.0.

Available Tools

6 tools
check_pricingB

Look up RunAPI pricing for the gemini-omni model line.

Inputs: model, action.

create_audioC

Run a synchronous Gemini Omni operation on RunAPI (create audio). Returns the operation result.

Inputs: name (required), audio_id (required), model, example_dialogue, +1 more.

create_characterC

Run a synchronous Gemini Omni operation on RunAPI (create character). Returns the operation result.

Inputs: descriptions (required), reference_image_url (required), model, audio_ids, +2 more.

get_taskB

Fetch the current status and latest result payload for a gemini-omni task.

Inputs: task_id (required), action.

loginA

Authenticate RunAPI by opening a browser PKCE login flow and saving the API key to ~/.config/runapi/config.json.

Inputs: force.

text_to_videoC

Create a Gemini Omni task on RunAPI (text to video). Returns a task id, status, and output URLs.

Inputs: prompt (required), seed, wait, model, +12 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    B3.2/5.0

    Scored across 6 tools

    Disambiguation4/5

    Each tool targets a fairly distinct action (auth, pricing lookup, task polling, and three generation endpoints), so an agent can mostly pick correctly. The main blur is between the synchronous create_character/create_audio operations and the task-based text_to_video + get_task flow, since it's unclear whether the sync creates also produce task ids that get_task can poll.

    Naming Consistency3/5

    Names mix three conventions: verb_noun (create_character, create_audio, get_task), verb_noun for a non-resource (check_pricing), a bare verb (login), and a noun-phrase endpoint name (text_to_video). It's still readable, but there is no single predictable pattern across all six.

    Tool Count5/5

    Six tools is well-scoped for a thin wrapper around one model line: auth, pricing, three generation entry points, and task polling. Nothing feels padded or redundant.

    Completeness3/5

    Core lifecycle is only partly covered: generation plus get_task exists, but there is no cancel_task, list_tasks, or bulk status retrieval, and login has no matching logout/config-status tool. The asymmetric model (two sync ops vs one async op) also leaves unclear how results of the sync operations are re-fetched.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Resources

    Related MCP Connectors

    Related MCP Servers