Orcarouter

Local

orcarouter.aiAI & Agents13Updated 22w ago

About Orcarouter

OrcaRouter’s official MCP server lets MCP clients browse its model catalog and run chat completions with selected models. Catalog details are available without an API key; chat calls require one.

Key features

01Browse provider and model catalogs
02Filter by provider, capability, or context
03Inspect model pricing, latency, and endpoints
04Run chat completions
05Auto-route requests and configure fallback chains

Use cases

01Compare model pricing and capabilities
02Inspect details for a specific model
03Run chat requests through automatic routing
04Set fallback models for chat requests

MCP support

The server provides MCP tools for browsing model and provider catalogs and running chat completions. Catalog tools need no API key; the chat tool requires an OrcaRouter API key.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config

{
  "mcpServers": {
    "orcarouter": {
      "command": "npx",
      "args": [
        "-y",
        "@orcarouter/mcp"
      ],
      "env": {
        "ORCAROUTER_API_KEY": "<ORCAROUTER_API_KEY>",
        "ORCAROUTER_BASE_URL": "<ORCAROUTER_BASE_URL>",
        "ORCAROUTER_REQUEST_TIMEOUT": "<ORCAROUTER_REQUEST_TIMEOUT>"
      }
    }
  }
}

Run commandnpx -y @orcarouter/mcp

Environment variables to set

ORCAROUTER_API_KEYSecretOrcaRouter API key. Required only for orcarouter_chat. Catalog tools (orcarouter_models_list, orcarouter_model_card, orcarouter_providers_list) work without it.

ORCAROUTER_BASE_URLOverride the gateway base URL. Defaults to https://api.orcarouter.ai.

ORCAROUTER_REQUEST_TIMEOUTPer-request timeout in seconds. Defaults to 300 (5 minutes) to accommodate long reasoning-model completions.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages

npm@orcarouter/mcp1.1.5stdio

Tools

Saved tool list · updated 3w ago

Parameter	Type	Required	Description
model	string	No	Model to call. Defaults to `orcarouter/auto` — your workspace's seeded auto-router. Use `orcarouter/<name>` for other workspace routers, or `<provider>/<model>` for direct upstream selection (e.g. `openai/gpt-4o-mini`, `anthropic/claude-haiku-4.5`).
models	array	No	Optional fallback chain. Models are tried in order if the primary fails. Max 5 entries including the primary.
prompt	string	Yes	User message to send (single-turn).
max_tokens	integer	No	Maximum tokens to generate (default 10000). Automatically translated to max_completion_tokens for OpenAI reasoning models.
temperature	number	No	Sampling temperature (default 0.7).
system_prompt	string	No	Optional system prompt prepended to the conversation.

Details

From the project README

OrcaRouter MCP Server

Official MCP server for the OrcaRouter LLM gateway.

Browse OrcaRouter's model catalog and run chat completions from inside any Model Context Protocol client — Claude Desktop, Claude Code, Cursor, Windsurf, Zed, or anything else that speaks the protocol.

Catalog browsing works without an API key — compare pricing and capabilities before signing up.

What you can do

🗺️ Discover providers and models without an API key
💬 Run chat completions through any served model
🧠 Auto-route requests through your workspace's orcarouter/auto router (cost / quality / balanced / LinUCB / gated-adaptive strategies)
🔁 Configure fallback chains (primary + up to 4 fallbacks) for resilience
📊 Filter models server-side by provider, capability, or minimum context window
🎯 Inspect detailed model cards: pricing, context, latency, supported endpoints
🔌 Works with Claude Desktop, Claude Code, Cursor, Windsurf, Zed, and any MCP client

Examples

Try saying things like:

"List all providers on OrcaRouter"
"Show me all Anthropic models with their pricing"
"Get details about minimax/minimax-m2.7"
"Chat with orcarouter/auto and explain quantum computing"

Quick Start

Claude Code (CLI one-liner)

claude mcp add orcarouter -s user \
  -e ORCAROUTER_API_KEY=sk-orca-your-key \
  -- npx -y @orcarouter/mcp

Other clients (config-file path)

Copy the example config for your MCP client into the client's config file:

Client Example Action
Claude Desktop claude-desktop.json Replace
Claude Code claude-code.json Merge
Cursor cursor.json Replace
Windsurf windsurf.json Replace

See examples/README.md for the config-file paths and notes on Zed and other clients.
Replace sk-or-... in the copied file with your OrcaRouter API key.
Restart your MCP client.

The root .mcp.json is the same config in the Open Plugins standard location, so registry/discovery tools that scan for it (e.g. cursor.directory) can pick this server up automatically.

Requires Node.js 18 or later. The ORCAROUTER_API_KEY env var is only required for orcarouter_chat; catalog tools work without it.

Tools

orcarouter_chat — run a chat completion (with optional fallback chain)
orcarouter_models_list — browse the catalog (pricing, context, capabilities)
orcarouter_model_card — detailed info for one model
orcarouter_providers_list — list providers with model counts

Full input schemas are exposed at runtime via the MCP tools/list method — your MCP client (Claude Desktop, Cursor, etc.) reads them automatically.

Configuration

Name	Required	Description
`ORCAROUTER_API_KEY`	optional	OrcaRouter API key. Required only for `orcarouter_chat`.
`ORCAROUTER_BASE_URL`	optional	API base URL. Defaults to `https://api.orcarouter.ai`.
`ORCAROUTER_REQUEST_TIMEOUT`	optional	Per-request HTTP timeout in seconds. Defaults to `300`.

Security

API keys are read from environment variables, never logged, and only sent to the OrcaRouter API. See SECURITY.md for the vulnerability disclosure policy.

Development

# bun (preferred)
bun install
bun run test
bun run typecheck
bun run build

# or with npm
npm install
npm test
npm run typecheck
npm run build

The build produces an ESM bundle at dist/index.js with a #!/usr/bin/env node shebang, runnable as the orcarouter-mcp binary.

Contributing

See CONTRIBUTING.md. For newcomer-friendly tasks, browse the good first issue label.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

Read the getting-started guide

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.

Client	Example	Action
Claude Desktop	claude-desktop.json	Replace
Claude Code	claude-code.json	Merge
Cursor	cursor.json	Replace
Windsurf	windsurf.json	Replace

Orcarouter

About Orcarouter

Key features

Use cases

MCP support

Configuration

Tools

Details

OrcaRouter MCP Server

What you can do

Examples

Quick Start

Claude Code (CLI one-liner)

Other clients (config-file path)

Tools

Configuration

Security

Development

Contributing

License

Comments

New to MCP?

What is MCP (Model Context Protocol)?

How do MCP servers differ from clients?

Should I use a local or a remote server?

How do I install a server in Claude, Cursor and others?

Are listed projects reviewed? What does “Verified” mean?

My project is listed. How do I claim it?