v0.1.1 - Local Admin router config

Highlights

• Added a local Admin page at /admin for runtime routing configuration.
• Added manual router-model selection for auto, auto-coding, and auto-longtext.
• Added local .llm-router.local.json runtime config persistence.
• Added protected /admin/config API for reading and saving router settings.
• Kept the OpenAI-compatible /v1/chat/completions behavior unchanged.

Validation

• npm run check passed locally.
• 24/24 tests passed, including Admin API auth, manual router-model routing, streaming, tools, multimodal routing, and fallback coverage.
• GitHub Actions CI passed on main.

Upgrade

Pull latest main, run npm install, then start the service and open http://127.0.0.1:8787/admin. If ROUTER_API_KEY is set, enter that local key in the Admin page.

v0.1.0 - OpenAI-compatible auto model router

Highlights

• OpenAI-compatible local router for one existing upstream relay key.
• Virtual models: auto, auto-coding, and auto-longtext.
• Non-streaming and streaming Chat Completions support.
• Tool calling passthrough and multimodal payload handling.
• Cheapest known-price router model selection and auto fallback.
• Public launch polish: CI, Docker, docs, examples, security policy, contribution guide, issue templates, and MIT license.

Validation

• npm run check passed locally.
• GitHub Actions CI passed on main.