aibotgo provides a reverse-engineered API gateway for various large language models, primarily focusing on enabling free access to long-text processing capabilities. It targets users who need to interact with models like Kimi AI, ChatGPT, Gemini, and others for tasks such as text summarization, intelligent conversation, web search, and OCR, offering a zero-configuration deployment and compatibility with OpenAI's API structure.

How It Works

The project acts as a proxy, intercepting requests and forwarding them to the respective AI services using extracted refresh_token credentials. It supports streaming output, multi-turn conversations, web search integration, and document/image analysis by leveraging the underlying models' capabilities. The design prioritizes ease of use and broad model compatibility, abstracting away the complexities of individual API integrations.

Quick Start & Requirements

• Installation: Docker (docker run -it -d --init --name aibotgo -p 9999:9999 -e TZ=Asia/Shanghai shengdingbox/aibotgo:0.0.1) or Docker Compose.
• Prerequisites: A server with a public IP and open port 9999. Java environment for native deployment.
• Setup: Minimal setup time via Docker.
• Documentation: 接入准备, Docker部署

Highlighted Details

• Supports a wide array of AI models including Kimi, ChatGPT, Gemini, Claude, and more.
• Offers OpenAI-compatible API endpoints (/v1/chat/completions).
• Features include long document analysis and image OCR.
• Supports multi-account access by concatenating refresh_tokens.

Maintenance & Community

The project is community-driven with no explicit mention of core maintainers or sponsorships. Links to community resources like Discord/Slack are not provided in the README.

Licensing & Compatibility

The project is presented as being for "research exchange and learning purposes only." It explicitly states: "For self-use only, prohibited from providing services to others or commercial use." This implies a non-commercial license, though no specific license is stated.

Limitations & Caveats

Reverse-engineered APIs are inherently unstable and subject to change or blocking by the service providers. The project warns against commercial use due to potential risks of account suspension. Multi-turn conversations are noted to be implemented via message merging, which may degrade performance and is subject to token limits.