OpenRouter API Configuration Guide

Overview

OpenRouter API is a unified interface platform that aggregates multiple LLM services, supporting access to models from OpenAI, Anthropic, Google, Meta, and more.

With OpenRouter, you can:

• Use a unified API to call models from different providers
• Auto-route to optimal models or lowest-cost nodes
• Avoid rate limiting issues from single platforms
• Switch models flexibly without changing code

Popular available models:

Model	Provider	Features	Use Cases
openai/gpt-4o-mini	OpenAI	Fast, low cost	Daily conversations
openai/gpt-4o	OpenAI	High quality	Multimodal tasks
anthropic/claude-3-haiku	Anthropic	Fast	Light tasks
anthropic/claude-3-opus	Anthropic	High reasoning	Complex analysis
google/gemini-pro	Google	Strong multimodal	General tasks

---

Getting API Key

Step 1: Visit OpenRouter

Go to OpenRouter

Step 2: Sign In

Supports GitHub / Google login.

Step 3: Create API Key

1. Navigate to Dashboard → Keys
2. Click Create Key
3. Copy and save your API Key

⚠️ Note: Keep your API Key secure and avoid exposing it.

---

Configuring PicoClaw

Option 1: Using WebUI (Recommended)

PicoClaw provides a WebUI interface where you can easily configure models without manually editing configuration files.

Edit preset settings, or click the "Add Model" button in the top right corner:

Field	Value
Model Alias	Custom name, e.g., gpt-4o-mini
Model Identifier	openai/gpt-4o-mini (or other supported models)
API Key	OpenRouter API Key
API Base URL	https://openrouter.ai/api/v1

Option 2: Edit Configuration File

Add in config.json:

{
  "model_list": [
    {
      "model_name": "gpt-4o-mini",
      "model": "openai/gpt-4o-mini",
      "base_url": "https://openrouter.ai/api/v1",
      "auth_method": "api_key",
      "api_key": "YOUR_OPENROUTER_API_KEY",
      "headers": {
        "HTTP-Referer": "http://localhost",
        "X-Title": "PicoClaw"
      }
    },
    {
      "model_name": "claude-3-haiku",
      "model": "anthropic/claude-3-haiku",
      "base_url": "https://openrouter.ai/api/v1",
      "auth_method": "api_key",
      "api_key": "YOUR_OPENROUTER_API_KEY"
    }
  ],
  "agents": {
    "defaults": {
      "model": "gpt-4o-mini"
    }
  }
}

---

Limits & Quotas

Billing

OpenRouter uses a pay-as-you-go model, charging based on the actual model used and token consumption.

Rate Limits

• Different models have different rate limits
• Free models may have stricter limits
• Paid users enjoy higher rate quotas

---

Common Issues

Model Unavailable

Cause: Model discontinued or insufficient account balance

Solutions:

• Check if the model is still available
• Top up account balance

Response Timeout

Cause: Slow model response or network issues

Solutions:

• Try using a faster model
• Check network connection