Open-weights s1-llm-auto-router picks the right LLM in 26ms on a CPU
airesearch12 · x · 2026-10-15
Decision Models released s1-llm-auto-router as open weights: a small decision model that reads a request and picks the right LLM for it, at a fraction of the cost of asking a big model.
- 26ms on a CPU, 22 languages, MIT license
- In a held-out routing test (2,175 synthetic requests, 22 languages, 5-LLM consensus labels) it hit 74.2% agreement with gold and 89.8% category accuracy, beating Jev 1.13, Weiche-395M and Winnow-12B
- msg group, a top-5 German IT consultancy (€1.5B revenue, 10,000+ employees), will evaluate it internally to make its LLM usage more efficient
- Live now on the Decision Models API; the company's first use-case-specific decision model
The thesis: replace big-model routing calls with tiny specialized decision models.
More from Infra
- Jensen Huang: Europe Has the Scientists but Needs a Lasting Energy Solution for AI — karmicoder · 2026-10-15
- Moving 1PB out of the cloud costs up to $74K in egress fees alone, $500K all-in — DavidLinthicum · 2026-10-15
- Robotics data gap: just 300K hours of manipulation data globally vs 100M+ needed, $3B spend projected — 0xSammy · 2026-10-15
- Ramp raises $1.85B at $60B valuation, says 85,000 firms save $1B+/month with AI tools — SethGRosenberg · 2026-10-15
- AI's smartest bear case is supply: $500B GPUs idle, but demand is elastic — bittingthembits · 2026-10-15
- Intel said to have secured foundry commitments from Broadcom and Google, Apple wins on 14A — BenBajarin · 2026-10-15