# workweave/router

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

Repository: https://github.com/workweave/router
Canonical: https://ross.abutalabs.com/products/workweave-router
Homepage: https://weaveos.com/products/router
Language: Go
License: NOASSERTION
License Family: other
Topics: agentic-coding, ai-gateway, anthropic, claude-code, codex, model-router, openai-compatible
Last push: 2026-08-26T16:34:43+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 60, longevity 9
- inputs: {"age_days": 128, "days_push": 7, "days_rel": 52, "gap_med": null, "n_releases_24m": 1}
- flags: prerelease_only, young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1103, forks 55 (observed 2026-08-28T04:03:22.738187+00:00)

## What it is
Workweave Router is a model router for agentic AI systems that routes each prompt to the most cost-effective model capable of handling it, claiming sub-50ms routing overhead and 40-70% cost savings. It exposes an OpenAI-compatible endpoint so tools like Claude Code, Codex, and Cursor can use it with just an endpoint change.

## Use cases
- cut llm inference costs for coding agents
- route prompts to cheaper models automatically
- use multiple llm models with claude code
- reduce token spend in cursor and codex
- openai-compatible llm proxy gateway
- smart model selection for agentic coding

## When to choose
- you use AI coding agents heavily and want to reduce frontier-model spend
- you want drop-in model routing without changing application code
- you need an OpenAI-compatible gateway that picks the cheapest sufficient model per request

## When to avoid
- you need full control over which model handles every request
- you require on-premises routing with no external service dependency
- your workload genuinely requires frontier models for nearly all prompts

## Facets
- artifact type: service
- maturity: active
- function: llm-inference, api-gateway, proxy, middleware
- domain: large-language-models, developer-tools, cloud-computing
- platform: go, self-hosted, cli
- tags: model-router, ai-gateway, openai-compatible, cost-optimization, claude-code, codex, agentic-coding, ai-agents, web-server

## Member repositories
- workweave/router (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:22.738187+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:45:34.628063+00:00, confidence not recorded.
  - readme: https://github.com/workweave/router (fetched 2026-08-28T04:03:22.738187+00:00, sha d726f127226e)
  - homepage: https://weaveos.com/products/router (fetched 2026-08-29T12:48:18.183933+00:00, sha 6fec51d84784)
  - site_page: https://weaveos.com/pricing (fetched 2026-08-29T12:48:18.193485+00:00, sha d20ba54b1ed7)
  - site_page: https://weaveos.com/product-pages/ai-agent (fetched 2026-08-29T12:48:18.195607+00:00, sha cc1b2281adcc)
  - site_page: https://weaveos.com/company (fetched 2026-08-29T12:48:18.197448+00:00, sha 3a315ea23e62)
- Data as of 2026-08-30T08:39:29.467469+00:00.