Decision Models' routing classifier hits AntSeed, cutting inference cost to 24% of always-Opus
airesearch12 · x · 2026-10-21
Decision Models is now live on AntSeed, an open marketplace for AI inference:
- Decision API: three decision models — s1-fast (fast/low-cost), s1-pro (highest accuracy), and s1-llm-auto-router (routing classifier) — outputting one answer per option with probabilities in a Jev-compatible format, plus a low-cost peer-to-peer tier and an EU privacy-focused tier.
- Auto Router: picks the best model per request across 176 models. Per the chart, always using Opus 5.5 for 60 requests costs 100%, while Auto Router scores 57/60 at just 24% of the cost.
A commercial example of small-model routing with big-model fallback cutting inference bills.
More from Infra
- Victor Taelin publicly begs to buy an 8-GPU Rubin machine — 'nobody is selling it' — fforres · 2026-10-21
- Nscale joins PyTorch Foundation as Platinum member after Anyscale acquisition move — PyTorch · 2026-10-21
- Decision Models launches on Antseed with an open-source auto model router — airesearch12 · 2026-10-21
- Gradium open-sources Phonon: 31M/40M edge TTS models with Rust inference stack for 10+ backends — l-m-z · 2026-10-21
- PostgreSQL 19 slips to November as LLM-assisted code review floods committers with edge cases — sanjaykalra · 2026-10-21
- Eon cuts engineer onboarding from 2 days to 2 minutes with an MCP Gateway on Bedrock AgentCore — AWS ML Blog · 2026-10-20