# opencode-llm-proxy

An OpenCode plugin that runs a local HTTP gateway translating between the OpenAI, Anthropic and Gemini API formats and whatever LLM provider OpenCode has configured, including streaming and parallel tool calling, so other tools only need one local endpoint.

opencode-llm-proxy starts a local server that accepts requests in the OpenAI Chat Completions format, the OpenAI Responses format, the Anthropic Messages format or the Gemini generateContent format, and forwards them through OpenCode to whichever provider OpenCode has been configured with, such as GitHub Copilot, Anthropic, Gemini, Ollama, OpenRouter or Amazon Bedrock. Tool and function calls, including several tool calls issued in a single turn, are translated between formats in both directions.

It has been used with clients such as n8n's AI agent node, Open WebUI, LangChain, the OpenAI and Anthropic SDKs, Gemini clients, and editor integrations such as Continue and Zed, letting each one reach any model OpenCode supports without being reconfigured separately.

It is written in JavaScript, distributed as an OpenCode plugin, and licensed under MIT.

Website: https://github.com/KochC/opencode-llm-proxy