Weave Router - Open-source model routing for coding agents

Show HN: Open-source model routing for coding agents at Astra-level performance

Weave Router is an open-source model router that plugs into any coding agent, like Claude Code or Codex, and intelligently switches between LLMs to match the task. It uses an ensemble approach to deliver frontier-level performance at a fraction of the cost. Weave Router 2.0 matches GPT-6 Astra on Terminal Bench 4.0 and SWE Atlas, while cutting costs by nearly half and running tasks over twice as fast. Built with a hidden Markov model and classifier for smart routing, it's available on GitHub or as a hosted service.

Matching frontier model performance was a huge milestone for our routing model, and in my opinion validates our initial hypothesis that an ensemble of models can do better than any single model ever could.
  1. creativeSlumber

    > then there are 10^100 possible paths through that session.

    Is this correct? in practice you would only be deciding what the next turn is going to be. Because the turn after the next turn is determined by the outcome of the next turn. So shouldn't this number be 10 * 100?

  2. YuechenLi

    Have you compared this to using GPT-6.1 Sol instead of GPT 6 Astra + Deepseek? From my test, 6.1 Sol is a lot more token efficient than 6 Sol while being similar to Astra in performance, and I don't really find 6 Astra to be significantly better than 6/6.1 Sol for general coding as I feel 6 Astra is only noticeably better at spatial reasoning/vision compared to 6 Sol, and 6.1 Sol really closed the gap on that front.

  3. thefourthchime

    Interesting work, and thanks for describing how your router works internally. It's definitely a fascinating subject. How would you say this compares to Cursor's auto mode?

  4. 361994752

    Sounds very interesting. Do you have plan to support gh copilot as model provider?

  5. svnt

    If you are training on data labeled by frontier models, how do you expect to exceed the performance of frontier models, other than in the cost dimension by recognizing simpler problems and routing to cheaper models?

More from this day

2026-10-01