brief productsopen models
Weave open-sources a coding-agent router
Weave open-sourced a model router that points Claude Code, Codex, or Cursor at one endpoint and picks the cheapest capable model per prompt.
Weave open-sourced a model router for coding agents: point Claude Code, Codex, or Cursor at one OpenAI- or Anthropic-compatible endpoint and it routes each request to the cheapest capable model in under 50ms, escalating to frontier models only when needed. It claims 40 to 70 percent cost savings and runs as a self-hostable proxy.
sources 2 cited
1 github.com workweave/router: model router for agentic systems 2 hyper.ai Weave Router routes LLM prompts to the right models, cutting costs