Supported Providers

CCProxy supports multiple AI providers through two methods: dedicated transformers and OpenAI-compatible endpoints. This page provides an overview of all supported providers.

How CCProxy Works

CCProxy uses a transformer-based architecture to translate API requests between Claude Code (which uses Anthropic's API format) and various AI providers. Instead of implementing direct provider clients, CCProxy uses transformers that:

Receive requests in Anthropic's API format
Transform them to the target provider's format
Forward the transformed request to the provider
Transform the response back to Anthropic's format
Return the response to Claude Code

This architecture allows CCProxy to remain lightweight while supporting multiple providers.

Native Provider Support

These providers have dedicated transformer implementations:

Provider	Speed	Cost	Models	Use Case	Transformer Status
Anthropic	⚡⚡⚡	💰💰💰	4+	Best reasoning & coding	✅ Full transformer
OpenAI	⚡⚡	💰💰💰	10+	Industry standard	✅ Full transformer
Google Gemini	⚡⚡	💰💰	5+	Multimodal AI	✅ Transformer (limited tools)
DeepSeek	⚡⚡	💰💰	3+	Advanced reasoning	✅ Transformer (no tools)
OpenRouter	⚡⚡	💰💰	100+	Model diversity	✅ Pass-through transformer

OpenAI-Compatible Providers

These providers work through the OpenAI transformer using OpenAI-compatible endpoints:

Provider	Speed	Cost	Models	Use Case	Tool Support
Ollama	⚡⚡⚡	💰	20+	Local/private AI	✅ Model dependent
Groq	⚡⚡⚡⚡	💰💰	8+	Ultra-fast inference	✅ Full support

Note: OpenAI-compatible providers use "openai" as the provider name in configuration but with different API base URLs.

Quick Setup

CCProxy uses a JSON configuration file. Here's the minimal configuration for each provider:

json

{
  "providers": [{
    "name": "anthropic",
    "api_base_url": "https://api.anthropic.com/v1",
    "api_key": "sk-ant-your_key_here",
    "models": ["claude-sonnet-4-20250720", "claude-opus-4-20250720"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "anthropic",
      "model": "claude-sonnet-4-20250720"
    }
  }
}

json

{
  "providers": [{
    "name": "openrouter",
    "api_base_url": "https://openrouter.ai/api/v1",
    "api_key": "sk-or-v1-your_key_here",
    "models": ["anthropic/claude-3.5-sonnet", "google/gemini-pro-1.5"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "openrouter",
      "model": "anthropic/claude-3.5-sonnet"
    }
  }
}

json

{
  "providers": [{
    "name": "openai",
    "api_base_url": "https://api.openai.com/v1",
    "api_key": "sk-your_key_here",
    "models": ["gpt-4", "gpt-3.5-turbo"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "openai",
      "model": "gpt-4"
    }
  }
}

json

{
  "providers": [{
    "name": "gemini",
    "api_base_url": "https://generativelanguage.googleapis.com/v1",
    "api_key": "your_key_here",
    "models": ["gemini-1.5-pro", "gemini-1.5-flash"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "gemini",
      "model": "gemini-1.5-pro"
    }
  }
}

json

{
  "providers": [{
    "name": "deepseek",
    "api_base_url": "https://api.deepseek.com/v1",
    "api_key": "your_key_here",
    "models": ["deepseek-chat", "deepseek-reasoner", "deepseek-coder-v2"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "deepseek",
      "model": "deepseek-chat"
    },
    "think": {
      "provider": "deepseek",
      "model": "deepseek-reasoner"
    }
  }
}

json

{
  "providers": [{
    "name": "openai",
    "api_base_url": "http://localhost:11434/v1",
    "api_key": "ollama",
    "models": ["llama3.1", "codellama", "mistral"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "openai",
      "model": "llama3.1"
    }
  }
}

json

{
  "providers": [{
    "name": "openai",
    "api_base_url": "https://api.groq.com/openai/v1",
    "api_key": "gsk_your_groq_api_key",
    "models": ["llama-3.1-70b-versatile", "llama-3.1-8b-instant"],
    "enabled": true
  }],
  "routes": {
    "default": {
      "provider": "openai",
      "model": "llama-3.1-70b-versatile"
    }
  }
}

Important Note: In each configuration above:

The models array lists available models for validation
The routes section defines which model actually handles requests
You must define at least a default route for CCProxy to work

Provider Comparison

Performance & Speed

Cost Comparison

Provider	Free Tier	Pricing Model	Best For
Anthropic	❌	Pay-per-use	Complex reasoning & coding
DeepSeek	✅ Limited	Pay-per-use	Advanced reasoning
OpenRouter	✅ Limited	Pay-per-use	Model variety
OpenAI	✅ Limited	Pay-per-use	Enterprise reliability
Gemini	✅ Generous	Pay-per-use	Multimodal tasks

For current pricing information, visit each provider's official pricing page.

Model Capabilities

Provider	Text	Code	Function Calling	Vision	Reasoning	Claude Code Compatible
Anthropic	✅	✅	✅	✅	✅	✅ Full Support
OpenAI	✅	✅	✅	✅	✅	✅ Full Support
Gemini	✅	✅	⚠️	✅	✅	⚠️ Limited support
DeepSeek	✅	✅	❌	❌	✅✅	⚠️ Limited (no tools)
OpenRouter	✅	✅	✅*	✅*	✅	✅ Model dependent
Ollama	✅	✅	✅**	❌	✅	✅ Model dependent
Groq	✅	✅	✅	❌	✅	✅ Full Support

*Depends on specific model routed through OpenRouter **Depends on specific model (e.g., Llama 3.1 supports tools)

Important: Tool Calling Requirement

⚠️ Critical for Claude Code Users: You must select models that support tool calling or function calling capabilities, as Claude Code requires these features to operate correctly.

Provider Tool Support Status:

Anthropic: ✅ Full tool/function calling support - Recommended for Claude Code
OpenAI: ✅ Full function calling support - Recommended for Claude Code
OpenRouter: ✅ Supports tools (model-dependent) - Check specific model capabilities
DeepSeek: ❌ No tool support in transformer - NOT recommended for Claude Code
Google Gemini: ⚠️ Limited tool support - May have compatibility issues with Claude Code

For best Claude Code compatibility, use Anthropic or OpenAI providers with models that explicitly support function calling.

Provider Selection Guide

Choose Anthropic if you want:

🧠 Best-in-class reasoning capabilities
💻 Superior coding performance (SWE-bench 72.5%)
📚 Long context window support (200K tokens)
🤔 Advanced thinking mode for complex problems

Choose OpenRouter if you want:

🎯 Access to 100+ different models
🔄 Model routing and fallbacks
🧪 Experimentation with cutting-edge models
📈 Usage analytics

Choose OpenAI if you want:

🏭 Industry-standard models
🛠️ Extensive tooling and ecosystem
👁️ Advanced vision capabilities
🎯 Proven reliability

Choose Google Gemini if you want:

🎥 Advanced multimodal capabilities
🏗️ Google's latest technology
📊 Strong analytical capabilities
🔍 Integration with Google services

Choose DeepSeek if you want:

🧠 Exceptional reasoning capabilities (87.5% on AIME 2025)
🎯 45-50% less hallucination
💻 Specialized coding models
🔬 Advanced mathematical and scientific analysis

Choose Ollama if you want:

🔒 Complete privacy - all processing happens locally
💰 Zero API costs - just your electricity
🌐 Offline capability after model download
🏠 Full control over your AI infrastructure

Choose Groq if you want:

⚡ Ultra-fast inference (10-100x faster than GPUs)
🚀 Near-instant responses for real-time applications
🛠️ OpenAI-compatible API for easy integration
📊 High-throughput workloads

Model Selection Guidelines

When selecting models from any provider:

Verify Tool Calling Support: Ensure the model supports function calling/tool use
Check Current Availability: Model availability changes frequently
Review Pricing: Visit the provider's official pricing page for current rates
Test Performance: Different models excel at different tasks

For current model lists, capabilities, and pricing, always check the provider's official documentation.

Getting Started

Pick a provider based on your needs
Get an API key from the provider's website
Configure CCProxy with your chosen provider
Start coding with Claude Code!

Ready to dive deeper? Click on any provider above to see detailed setup instructions, model lists, and configuration options.

Supported Providers ​

How CCProxy Works ​

Native Provider Support ​

OpenAI-Compatible Providers ​

Quick Setup ​

Provider Comparison ​

Performance & Speed ​

Cost Comparison ​

Model Capabilities ​

Important: Tool Calling Requirement ​

Provider Tool Support Status: ​

Provider Selection Guide ​

Choose Anthropic if you want: ​

Choose OpenRouter if you want: ​

Choose OpenAI if you want: ​

Choose Google Gemini if you want: ​

Choose DeepSeek if you want: ​

Choose Ollama if you want: ​

Choose Groq if you want: ​

Model Selection Guidelines ​

Getting Started ​

Supported Providers

How CCProxy Works

Native Provider Support

OpenAI-Compatible Providers

Quick Setup

Provider Comparison

Performance & Speed

Cost Comparison

Model Capabilities

Important: Tool Calling Requirement

Provider Tool Support Status:

Provider Selection Guide

Choose Anthropic if you want:

Choose OpenRouter if you want:

Choose OpenAI if you want:

Choose Google Gemini if you want:

Choose DeepSeek if you want:

Choose Ollama if you want:

Choose Groq if you want:

Model Selection Guidelines

Getting Started