Prerequisites Before you begin, create an account and get your API Key.

​
Getting started

Connect to OpenRouter to access multiple AI models, compare model performance, and manage usage across different AI providers through AI agents.
1

Install the SDKs (optional)

pip install klavis
2

Create a server instance

from klavis import Klavis
from klavis.types import McpServerName

klavis_client = Klavis(api_key="YOUR_API_KEY")

# Create an OpenRouter MCP server instance
openrouter_server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.OPENROUTER,
    user_id="user123"
)
Response Information: The API returns:
  • serverUrl: The URL for connecting your MCP client to OpenRouter
  • instanceId: Unique identifier for your server instance
  • Note: OpenRouter requires an API key which should be configured after creation
3

Configure authentication

You can get your OpenRouter API key from the OpenRouter website.
# Set the OpenRouter API key for your instance
response = klavis_client.mcp_server.set_instance_auth(
    instance_id=openrouter_server.instance_id,
    auth_data={
        "token": "YOUR_OPENROUTER_API_KEY"
    }
)
🎉 Your OpenRouter MCP Server is ready! You can now use your MCP server URL with any MCP-compatible client to access multiple AI models.

​
Available Tools

Tool NameDescription
openrouter_list_modelsList available models on OpenRouter
openrouter_search_modelsSearch for models based on query, category, or provider
openrouter_get_model_pricingGet pricing information for a specific model
openrouter_create_chat_completionCreate a chat completion using a selected model
openrouter_create_chat_completion_streamCreate a streaming chat completion
openrouter_create_completionCreate a text completion (legacy completion endpoint)
openrouter_get_usageGet usage statistics for the authenticated user
openrouter_get_user_profileGet the current user’s profile information
openrouter_get_creditsGet the current user’s credit balance
openrouter_get_api_key_infoGet information about the current API key
openrouter_get_cost_estimateEstimate cost for a model given input/output tokens
openrouter_compare_modelsRun the same prompt across multiple models and compare
openrouter_analyze_model_performanceAnalyze one model over multiple prompts
openrouter_get_model_recommendationsGet model recommendations based on use case and constraints
For more details about tool input schema, use the get_tools API.

​
Next Steps

AI Platform Integrations

Integrate Klavis MCP Servers with leading AI platforms

Integrations

Explore available MCP servers

API Reference

REST endpoints and schemas