# Ofox AI

> LLM Gateway + RAG + Agent Apps + Integrations. One-stop platform helping developers rapidly build AI Agent applications.

Ofox AI is a unified API gateway that provides access to 50+ top AI models (GPT, Claude, Gemini, DeepSeek, Qwen, etc.) through a single API endpoint. We offer competitive pricing, enterprise-grade reliability, and China-optimized infrastructure.

## Core Features

- **Unified API**: Access OpenAI, Anthropic, Google, and other providers through one API
- **Best Prices**: Up to 30% off on flagship models, 50% off on open-source models
- **High Availability**: 99.9% uptime SLA with global CDN acceleration
- **China Optimized**: Dedicated China endpoints with Aliyun/Volcengine acceleration
- **RAG Support**: Built-in knowledge base and retrieval-augmented generation
- **MCP Integration**: Model Context Protocol support for agent applications

## API Endpoints

### Global
- OpenAI Compatible: `https://api.ofox.ai/v1`
- Anthropic Compatible: `https://api.ofox.ai/anthropic`
- Gemini Compatible: `https://api.ofox.ai/gemini`

### China (Optimized)
- OpenAI Compatible: `https://api.ofox.ai/v1`
- Anthropic Compatible: `https://api.ofox.ai/anthropic`
- Gemini Compatible: `https://api.ofox.ai/gemini`

## Pages

- [Home](https://ofox.ai): Main landing page with product overview
- [Pricing](https://ofox.ai/pricing): Pricing plans (Free, Pro, Enterprise)
- [Vibe Coding](https://ofox.ai/vibe-coding): Unlimited Claude for Claude Code, Cursor, OpenCode, Cline developers
- [China Zone](https://ofox.ai/zh/china): China-specific features and endpoints

## Vibe Coding - Unlimited Claude for AI Coding Tools

> For developers using Claude Code, Cursor, OpenCode, or Cline who hit rate limits.

### Problem Solved
- Claude API rate limits interrupting your coding flow
- 4xx/429 errors breaking your vibe coding session
- High latency for developers in China

### Solution
Change one environment variable, get unlimited Claude:

```bash
export ANTHROPIC_BASE_URL=https://api.ofox.ai/anthropic
export ANTHROPIC_API_KEY=<YOUR_OFOXAI_KEY>
```

### Benefits
- **Unlimited RPM/TPM**: No more rate limits
- **Native Anthropic Protocol**: Zero code changes, full compatibility
- **99.9% SLA**: Enterprise-grade reliability
- **China Optimized**: <100ms latency with api.ofox.ai
- **Works with**: Claude Code, Cursor, OpenCode, Cline, Windsurf, Aider

### Keywords
claude code rate limit, cursor api alternative, claude api unlimited, vibe coding, ai coding tools, anthropic api proxy, claude china, cursor 国内加速

## Quick Start (Python)

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.ofox.ai/v1",
    api_key="<YOUR_API_KEY>"
)

response = client.chat.completions.create(
    model="openai/gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)
```

## Supported Models

### Flagship Models
- OpenAI: GPT-4o, GPT-4 Turbo, o1, o3
- Anthropic: Claude 3.5 Sonnet, Claude 3 Opus
- Google: Gemini 2.0 Flash, Gemini 1.5 Pro
- xAI: Grok-2

### Open Source Models
- DeepSeek: DeepSeek-V3, DeepSeek-R1
- Alibaba: Qwen-2.5, Qwen-Max
- Meta: Llama 3.3
- Mistral: Mistral Large

## Contact

- Website: https://ofox.ai
- GitHub: https://github.com/ofoxai
- Twitter: https://x.com/ofoxai

## Optional

- robots.txt: https://ofox.ai/robots.txt
- sitemap.xml: https://ofox.ai/sitemap.xml