Strot - the open decision compiler
Open-source decision compiler: a frontier LLM is used once at compile time to turn an intent plus schema into a frozen Python artifact, then every record is decided locally by a 144M model as typed JSON with confidence scores - low-confidence records abstain instead of guessing.
Why it matters
Agents at compile time, never at runtime - zero marginal inference cost and bounded outputs for classification, routing, and scoring workloads.
What you could build with it
A data team could compile a support-ticket triage classifier once and run it locally on thousands of records a day for near-zero marginal cost, with low-confidence cases routed to humans.
Does it hold up?
Too early to judge - launched hours ago. The only evidence is the author's own numbers (92.5% compiled vs 74.8% naive on 398 grounded-routing records, ~60ms per record, zero marginal cost, all in their own Docker setup) and one independent note that the Pac-Man demo exposes model-variance limits. No third-party deployments found yet.
Built with Strot
- STROT compiles LLM instructions into locally run decision workflowsruntimewire.com · Independent Oct 19 2026 writeup of the Show HN launch: describes the compile-once/decide-forever split and applies skeptical framing - the project's own Pac-Man demo shows the limits of that split and compilation plus golden tests cannot replace evaluation on real, changing data.
Learn more
independent technical analysis →
First spotted on github: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74Tencent Hunyuan Hy3: open 295B MoE with 21B active under Apache 2.0Tencent open-sourced Hy3, a 295B-parameter mixture-of-experts model with only 21B parameters active per token, under…models · JEV 0.73NVIDIA Isaac Lab 3.0 GA + Isaac OS open runtime + Agile ONE SN humanoidMajor Isaac platform expansion at GTC Berlin Oct 21: Isaac Lab 3.0 is generally available as the open-source robot…models · JEV 0.73StepFun Step-5-Preview — open BF16 weights releasedStepFun published the full BF16 weights for its flagship Step-5-Preview agentic model (600B-parameter sparse MoE, 27B…models · JEV 0.73NVIDIA Nemotron 3.5 Retriever collection: Embed 8B/1B, Rerank 8B, Parse 631MNVIDIA's Oct 21 collection drop for agentic retrieval: Nemotron-3.5-Embed in 8B BF16, 1B BF16 and 1B NVFP4…models · JEV 0.72Kimi K3.1 Pro: Moonshot's 2.8T open-weight flagship officially launchesMoonshot AI officially launched Kimi K3.1 Pro on Oct 23, a mixture-of-experts model with 2.8 trillion total / 104…models · JEV 0.72
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.