Back to directory

Weave Router

A model router that sits in front of Anthropic, OpenAI and Gemini APIs and picks the best available model for each request using a small on-device embedder rather than a prompt-based heuristic.

Weave Router

Weave Router acts as a drop-in proxy: pointing Claude Code, Codex, Cursor or a custom app at its local endpoint lets a cluster scorer, derived from Avengers-Pro, pick a model per action (rather than per whole conversation turn) from the providers a user has enabled.

It speaks the Anthropic Messages, OpenAI Chat Completions and native Gemini APIs, including streaming, tools and vision, and also routes to open models such as DeepSeek, Kimi, GLM, Qwen, Llama and Mistral through OpenRouter or another OpenAI-compatible endpoint. Provider keys stay on the user's machine, encrypted at rest, and it emits OpenTelemetry traces viewable in a bundled dashboard or an external tool such as Honeycomb, Datadog or Grafana.

It installs with a hosted quickstart command or self-hosts with Postgres via a make full-setup target; it is written in Go.

More in LLM gateways

See all 106
  • aihappey-ai

    github.com

    A stateless, multi-provider .NET AI gateway exposing normalized OpenAI- and Anthropic-compatible endpoints, managed agents and skills across a very large catalog of models.

  • Unified AI System

    happy520ai.github.io

    A self-hosted, protocol-first AI gateway and agent control plane for OpenAI, Anthropic, Gemini, MCP and A2A, turning a rough request into a structured, reviewable prompt before it runs.

  • Subroute

    github.com

    A local AI gateway and desktop control desk that gives coding tools one steady endpoint, so switching between AI subscriptions, API providers and local models doesn't mean reconfiguring every client.