duolahypercho/codex-router

Health v2 · maintenance only

Flags: young

Adoption not part of the score

2953 stars · 224 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A local model router that lets external LLM providers (Anthropic, Kimi, DeepSeek, xAI, GitHub Copilot, and others) be used inside the Codex App/CLI, DeepSeek Harness, and Gemini CLI through one credential-isolating gateway. It speaks the Responses API, merges routed models into native model pickers, and supports safe migration and rollback of configuration.

Use cases

• use DeepSeek or Kimi models inside Codex CLI
• route external LLM providers through one local gateway
• keep API keys isolated from chat agents
• merge third-party models into Codex's model picker
• safely migrate Codex config with rollback
• share one provider credential set across multiple AI coding clients

When to choose

• you use Codex, Gemini CLI, or DeepSeek Harness and want non-native models
• you want credential isolation and guided OAuth setup
• you need safe migration and rollback for model routing config

When to avoid

• you only use native OpenAI GPT models in Codex
• you need a hosted or multi-user routing service rather than a local one
• you want a general-purpose LLM proxy unrelated to Codex-style clients

Facets

cli-tool · maturity active

llm-inference api-gateway proxy configuration-management cli developer-tools large-language-models apis cli cross-platform codex model-router litellm deepseek kimi oauth model-catalog local-gateway command-line macos linux nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="duolahypercho/codex-router")