Skip to content

Show HN: Smart model routing directly in Claude, Codex and Cursor

8.4 relevance
Score Breakdown
technical depth
9
novelty
8
actionability
9
community
7
strategic
6
personal
10

Scored daily by a customisable AI persona to surface the most relevant engineering leadership news.

Smart model routing for coding agents is directly applicable to the reader's workflow and agent orchestration interests.

AI/ML github.com
Model router for agentic systems. Routes every prompt to the right model in <50ms. Cut costs 40-70% with just an endpoint change. - workweave/router
Summary

Weave Router is a drop-in localhost proxy that routes each request to the optimal model from Anthropic, OpenAI, Gemini, and open-source providers (DeepSeek, Llama via OpenRouter) using an on-box embedder based on the Avengers-Pro cluster scorer, not prompt-based heuristics. It supports streaming, tools, and vision across all three major API formats, with BYOK encryption and built-in OTLP observability to Weave, Honeycomb, or Grafana. A single `npx @workweave/router` command wires Claude Code, Codex, Cursor, or opencode in 30 seconds, with self-hosted Postgres-backed deployment as an alternative.

Author

workweave