Duck2api provides a unified API endpoint for various large language models, acting as a proxy to simplify integration for developers. It supports models like Claude-3-Haiku, Llama-3.3-70b, Mixtral-8x7b, and GPT-4o-mini, offering a single interface to interact with different LLM providers.

How It Works

Duck2api functions as a reverse proxy, forwarding requests to specified LLM APIs. It abstracts away the complexities of individual model integrations, presenting a consistent API structure. This approach allows users to switch between different LLM backends without modifying their client applications, promoting flexibility and reducing vendor lock-in.

Quick Start & Requirements

• Install/Run:
  • Binary: git clone https://github.com/aurora-develop/duck2api && cd duck2api && go build -o duck2api && chmod +x ./duck2api && ./duck2api
  • Docker: docker run -d --name duck2api -p 8080:8080 ghcr.io/aurora-develop/duck2api:latest
  • Docker Compose: Download docker-compose.yml and run docker-compose up -d.
• Prerequisites: Docker and Docker Compose for containerized deployments.
• Usage: curl --location 'http://<your-server-ip>:8080/v1/chat/completions' --header 'Content-Type: application/json' --data '{ "model": "gpt-4o-mini", "messages": [{"role": "user", "content": "Say this is a test!"}], "stream": true }'
• Docs: Web UI available at http://<your-server-ip>:8080/web.

Highlighted Details

• Supports multiple LLMs: Claude-3-Haiku, Llama-3.3-70b, Mixtral-8x7b, GPT-4o-mini.
• Provides a unified /v1/chat/completions API endpoint.
• Configurable via environment variables for authorization, TLS, and proxy settings.

Maintenance & Community

• Active pull request contributions are acknowledged.
• No specific community channels (Discord/Slack) or roadmap links are provided in the README.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive MIT license allows for commercial use and integration into closed-source projects.

Limitations & Caveats

The project mentions that gpt-3.5-turbo is no longer supported due to the DuckDuckGo API removing support. The README does not detail specific performance benchmarks or advanced configuration options beyond environment variables.