Vox MCP: Multi-Model AI Gateway

The right model for the right question beats the best model for every question.

Vox MCP is a multi-model AI gateway that lets you access any AI provider directly from Claude Code, Claude Desktop, Cursor, or any MCP client. Unlike other multi-model tools, Vox uses a pure passthrough design: prompts go to providers unmodified and responses come back unmodified. No system prompt injection, no response formatting, no behavioral directives.

Current release: 0.9.0 (opens in a new tab). Install the published package with uvx vox-mcp or pin it with uvx vox-mcp@0.9.0. Check saved model names after upgrading. listmodels reports configured choices, not provider liveness: some configured IDs may be retired or absent from their gateway catalogue. Check the actual provider or gateway lifecycle notice before reusing a saved selection.

Why it matters

When you're working in Claude Code and want a second opinion from Gemini, GPT, GLM, Kimi, or DeepSeek, you'd normally have to switch applications. Vox lets you query any model without leaving your current workflow.

The difference from alternatives: Most multi-model tools inject their own system prompts or modify your messages. Vox doesn't. What you send is what the model receives.

Supported providers

You only need API keys for providers you want to use. Vox works with any subset.

Tools

Vox provides three tools through the MCP protocol:

chat

Send prompts to any supported model with optional file or image attachments.

    "Use vox chat with gemini-3.8-flash:
Compare these two theoretical frameworks and identify tensions..."

listmodels

Show the configured catalog, aliases, capabilities, and allowlist restrictions for that Vox process. Provider configuration and local registries decide this output; it does not validate credentials, billing, or model entitlement. Gateway availability is not probed. Check the provider or gateway's documentation for the exact route, and verify account access separately before relying on it.

dump_threads

Export conversation threads as JSON or Markdown. This is useful for documenting multi-model analysis.

Multi-turn conversations

Vox supports persistent threads via continuation_id. This means you can:

1. Start a conversation with Gemini about a theoretical framework
2. Continue the same thread with follow-up questions
3. Switch to DeepSeek mid-conversation to get a different perspective
4. Export the entire multi-model dialogue

Vox automatically saves threads to disk as JSONL for durability. You can export threads as Markdown.

Research workflows

Host support

Vox is a standard stdio MCP server, so it runs anywhere MCP runs.

The exact config blocks for the common hosts live in the Installation section.

Install

Two ways to set Vox up:

• Published package — use uvx vox-mcp in the MCP client configuration below; pin vox-mcp@0.9.0 when reproducibility matters.
• Manually — clone, configure .env, register with your MCP client. See the steps below.
• Via Carrel — run /carrel-setup and answer yes when the interview asks about multi-model access (Gemini, GPT, Grok, Kimi, …). Carrel adds Vox at project level and prompts for at least one provider API key.

The manual path is host-agnostic; the Carrel path is Claude Code-only but skips the config steps.

Setup

git clone https://github.com/linxule/vox-mcp.git
cd vox-mcp
uv sync

cp .env.example .env
# Edit .env — add at least one provider API key

uv run python server.py

For a published-package configuration, use uvx instead of a source checkout:

{
  "mcpServers": {
    "vox-mcp": {
      "command": "uvx",
      "args": ["vox-mcp@0.9.0"],
      "env": {
        "OPENAI_API_KEY": "your-key-here"
      }
    }
  }
}

MCP client configuration

Vox runs as a stdio MCP server. Replace /path/to/vox-mcp with the absolute path to your cloned repo.

claude mcp add vox-mcp \
  -e GEMINI_API_KEY=your-key-here \
  -- uv run --directory /path/to/vox-mcp python server.py

{
  "mcpServers": {
    "vox-mcp": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/vox-mcp", "python", "server.py"],
      "env": {
        "GEMINI_API_KEY": "your-key-here"
      }
    }
  }
}

{
  "mcpServers": {
    "vox-mcp": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/vox-mcp", "python", "server.py"],
      "env": {
        "GEMINI_API_KEY": "your-key-here"
      }
    }
  }
}

{
  "mcpServers": {
    "vox-mcp": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/vox-mcp", "python", "server.py"],
      "env": {
        "GEMINI_API_KEY": "your-key-here"
      }
    }
  }
}

{
  "mcpServers": {
    "vox-mcp": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/vox-mcp", "python", "server.py"],
      "env": {
        "GEMINI_API_KEY": "your-key-here"
      }
    }
  }
}

Configuration options

Beyond API keys, Vox supports several configuration options in .env; the repository's .env.example is the full reference:

Saved defaults and gateway routes

Vox 0.9 retains the ability to persist a model preference without editing the MCP client file:

uvx vox-mcp config set-default gpt-6-astra
uvx vox-mcp config show

Use listmodels in the connected client to see the model catalog configured for that process. A saved default is a preference, not a provider-access check; explicit chat.model values and continued-thread models take precedence.

Restart or reconnect Vox after changing the saved default. A DEFAULT_MODEL value in the MCP server process environment overrides the saved setting. config show reports its own launch environment, which may differ from the MCP client's.

Cloudflare and Vercel AI Gateway routes use an explicit first prefix, for example cloudflare/openai/gpt-5.5 or vercel/anthropic/claude-sonnet-4.6. Vox keeps the prefix in conversation memory, refuses an unconfigured or disallowed gateway, and does not silently fall through to another provider. Gateway requests currently support text only; native-provider IDs and gateway IDs remain distinct.

Provider behavior notes

• Vox routes Gemini requests through Google's Interactions API when available, using it statelessly (store=false) so Vox still owns conversation memory. Image inputs and failed Interactions attempts fall back to generateContent; set VOX_GEMINI_USE_INTERACTIONS=false to skip the Interactions path.
• Gemini thinking controls differ by API surface. Gemini 3 uses thinking_level; Gemini 2.x still has thinking support, but the Interactions path maps Vox's thinking_mode to the levels that API accepts.
• Vox omits temperature unless the caller explicitly supplies one. That preserves provider defaults for models where changing temperature is discouraged or unsupported.

Part of Research Memex

Vox integrates naturally with other Research Memex tools:

• Interpretive Orchestration Plugin — Multi-model triangulation during qualitative analysis
• Kimi Plugin for Claude Code — For full delegation/review jobs rather than one-shot consults
• Claude Code Setup Guide — Your primary research environment
• AI Model Reference Guide — Understanding which models to query for what