switchboard
W

Workweave Router

Verified

by Weave

Routes each prompt to the optimal model (Anthropic, OpenAI, Gemini) in under 50ms via an on-box embedder. Router 2.0 adds subscription-aware routing for Claude Code, Codex, and opencode.

Access
CLIAPI
Auth
API Key
Skills
3

Skills

Sub-50ms Routing

Routes each prompt to the best model in under 50ms using an on-box embedder, no extra network hop.

Drop-In Proxy

Acts as an Anthropic- and OpenAI-compatible proxy for Claude Code, Codex, and opencode.

Cost Reduction

Cuts model spend 40-70% by sending simple prompts to cheaper models and hard ones to frontier models.

Infrastructure & Opsmodel-routingproxycost-optimizationclaude-codecodexopen-sourcellm-gateway
Visit Agent

Related Agents

OpenRouter

by OpenRouter

VerifiedOfficial

Unified API gateway routing requests across hundreds of language models from many providers, with automatic fallback…

Language & TranslationInfrastructure & Ops
API
API Key

Headroom

by Headroom Labs

Verified

Compresses tool outputs, logs, RAG chunks, files, and history locally before they reach the model, as a Python or…

Infrastructure & OpsCode & DevTools
APIMCPCLI
5 skills · No auth

LiteLLM

by BerriAI

Verified

AI gateway unifying 100+ LLM providers under one OpenAI-compatible API. Cost tracking, guardrails, load balancing. pip…

Infrastructure & Ops
APICLI
3 skills · API Key

Portkey

by Portkey

Verified

AI gateway with unified API for 200+ LLMs, plus observability, caching, fallbacks, and cost management. Drop-in…

Infrastructure & Ops
API
5 skills · API Key
workweave-router
Routes each prompt to the optimal model (Anthropic, OpenAI, Gemini) in under 50ms via an on-box embedder. Router 2.0 adds subscription-aware routing for Claude Code, Codex, and opencode.
fields
nameWorkweave Router
providerWeave
urlhttps://github.com/weave-os/router
categoriesinfrastructure
accesscli · api
authapiKey
streamingfalse
pushfalse
verifiedtrue
officialfalse
tagsmodel-routing, proxy, cost-optimization, claude-code, codex, open-source, llm-gateway
skills
sub-50ms-routingSub-50ms RoutingRoutes each prompt to the best model in under 50ms usin…
drop-in-proxyDrop-In ProxyActs as an Anthropic- and OpenAI-compatible proxy for C…
cost-reductionCost ReductionCuts model spend 40-70% by sending simple prompts to ch…
→ https://github.com/weave-os/router