{"type":"mcp_server","name":"GPT_API_free","description":"This repository provides a free API for accessing various large language models like GPT, DeepSeek, Claude, Gemini, and Grok, offering a unified protocol for easy integration and domestic acceleration without requiring a proxy.","category":"AI","language":"Python","stars":38448,"forks":2222,"owner":"chatanywhere","github_url":"https://github.com/chatanywhere/GPT_API_free","homepage":"https://api.chatanywhere.tech","setup":"- **[Apply for Internal Beta Free API Key](https://api.chatanywhere.org/v1/oauth/free/render)**\n- The free version supports deepseek, gpt-3.5-turbo, embedding, gpt-4o-mini, gpt-4o.\n- **Forward Host1: `https://api.chatanywhere.tech` (Domestic relay, lower latency)**\n- **Forward Host2: `https://api.chatanywhere.org` (Overseas use)**","tools":"1.  Supports popular large models such as gpt | deepseek | claude | gemini | grok.\n2.  The free version supports gpt-4o, gpt-4.1 5 times a day; supports deepseek-r1, deepseek-v3 30 times a day, supports gpt-4o-mini, gpt-3.5-turbo, gpt-4.1-mini, gpt-4.1-nano 200 times a day.\n3.  Interface standards that are completely consistent with the official ones, compatible with various software/plugins.\n4.  Supports streaming responses.\n5.  Domestic lines use dynamic acceleration, and the experience is far better than using a proxy to connect to the official website.\n6.  No need for scientific internet access, it can be used directly in the domestic environment.\n7.  Completely free for personal use.\n8.  The protocol uniformly uses the openai standard protocol. For other manufacturer models, only the model name needs to be changed for easy access.","faq":null,"created_at":"2023-04-23T12:42:08+00:00","updated_at":"2025-07-07T16:33:06+00:00","source_url":"https://model-context-protocol.com/servers/gpt-api-free","related_articles":[{"title":"GPT_API_free: Complete Implementation Guide","url":"https://model-context-protocol.com/blog/gptapifree-guide"}]}