Router.com
#6 en APIs & Gateways LLMRamp · depuis 19. August 2026 · 3× · vu le 24 août 2026
48
Momentum
Router.com is an API service developed by Ramp for routing LLM requests across multiple providers. A single endpoint automatically routes each request to the cheapest model that clears a set quality threshold, with automatic fallback on provider failures. The service is built on routing infrastructure Ramp has run internally in production for three years and launched publicly as an independent brand on August 19, 2026. Router.com supports models from providers including OpenAI, Anthropic, SpaceXAI (formerly xAI), DeepSeek, Moonshot, Minimax, Nvidia, and Z.ai, with more providers coming soon.
Historique du momentum
26.05.24.08.
Fonctionnalités
| Throughput/Latency | Over 100 optimizations across model selection, caching, compression, timing; 99.9%+ reliability in Ramp's internal production traffic |
| License | Proprietary Ramp service, usage subject to the Router Terms of Service |
| Platform | Cloud API at router.com / api.router.com, currently available in the US only |
| Price | Free through end of 2026 (users still pay list price for model tokens); $26 launch credit; no 2027 pricing announced yet |
| Protocol Compatibility | Own API compatible with OpenAI Responses format (endpoint api.router.com/v1/responses); reportedly also usable with Anthropic-format requests |
| Release Date | August 19, 2026 (public launch) |
| Supported Models/Providers | OpenAI, Anthropic, SpaceXAI (formerly xAI), DeepSeek, Moonshot, Minimax, Nvidia, Z.ai/GLM, Qwen (via Fireworks AI); Gemini, Google, AWS, Together AI, Baseten, Crusoe coming soon |