Codex (ChatGPT) Integration
Claude Code Router can use a ChatGPT/Codex subscription to route Claude Code requests through OpenAI's models. The Codex backend powers OpenAI's ChatGPT product and this integration lets you leverage that subscription with Claude Code.
Qwen Chat Integration
Claude Code Router supports routing through Qwen Chat (通义千问) via the qwen.aikit.club API. It uses JWT-based authentication where you paste a token copied from the chat.qwen.ai web application.
Chrome On-Device (Gemini Nano)
Claude Code Router can use Chrome's built-in Gemini Nano model — a ~4GB local LLM that runs entirely on your device with zero API cost and zero latency to external providers. This is accessed through Chrome's Prompt API via the Chrome DevTools Protocol (CDP).
DeepSeek Reasoning Replay
Some DeepSeek-based providers (e.g., via OpenCode, ZenGo) require previous assistant reasoning content to be included in subsequent requests. This is because DeepSeek splits its output into visible content and internal reasoning/chain-of-thought, and the reasoning must be fed back into the conversation context for the model to maintain coherence across turns.
Model Discovery
Claude Code Router can auto-discover available models from any API provider that exposes a model listing endpoint. This is useful for providers with frequently changing model lists, or for exploring what's available on a new provider.