Releases·Tool Release
Ollama 0.35 Adds /v1/systemone Endpoint for Local Nimble Decision Models
Ollama 0.35 introduces a /v1/systemone endpoint that lets agents run typed routing, moderation, and scoring decisions on local hardware, using Nimble decision models. It gives practitioners building local agent stacks a structured decision-making primitive without cloud calls.
0 engagement1 sourceFri, Oct 2, 2026, 05:06 AM
AI Overview
- Ollama 0.35 introduces a /v1/systemone endpoint letting agents run Jev-style decision models locally.
- The endpoint supports typed routing, moderation, and scoring decisions on local hardware.
- The release ships with Nimble, a decision-model setup the source says is the bigger change.
Entities
Related
Releases·Tool Release
Rizzo AI Academy releases Rizzo Flow, open-source local implementation of TypeSafe's Jev typed-decision model
Rizzo AI Academy released Rizzo Flow, an open-source, local-first implementation of the idea behind Jev, TypeSafe's 'System One' model. The tool takes unstructured state as input and produces typed, probabilistic decisions from an LLM without generating a single token, running entirely on the user's own machine. It offers developers a way to get structured decision outputs locally without token generation costs.
Rizzo AI Academy released Rizzo Flow, an open-source, local-first implementation of the idea behind Jev, TypeSafe's 'System One' model. The tool takes unstructured state as input and produces typed, probabilistic decisions from an LLM without generating a single token, running entirely on the user's own machine. It offers developers a way to get structured decision outputs locally without token generation costs.
240 engagement1 sourcegithubMon, Sep 21, 2026, 12:00 AM
Social·Community
System 1 Models: Faster and Cheaper Proxy to Frontier Models — With a Working Router
Two Medium posts examine TypeSafe's System One family of decision-focused models — Jev, Laya, Kev, and Nimble — covering their architecture, typed decisions, parallel inference, and calibration. A related post demonstrates routing requests through a LiteLLM proxy to pick the appropriate model. The posts position these open-weights models as faster, cheaper proxies to frontier models for decision workloads rather than conversation.
Two Medium posts examine TypeSafe's System One family of decision-focused models — Jev, Laya, Kev, and Nimble — covering their architecture, typed decisions, parallel inference, and calibration. A related post demonstrates routing requests through a LiteLLM proxy to pick the appropriate model. The posts position these open-weights models as faster, cheaper proxies to frontier models for decision workloads rather than conversation.
0300 engagement3 sourcesrssThu, Sep 24, 2026, 12:00 AM
Social·Community
From Generation to Decision: Why Jev Could Change How We Build LLM Applications
A JIN System Architect blog post argues Jev could change LLM application architecture by having AI output decisions software can compute, rather than generated text. The excerpt is too thin to state the mechanism or evidence in detail.
A JIN System Architect blog post argues Jev could change LLM application architecture by having AI output decisions software can compute, rather than generated text. The excerpt is too thin to state the mechanism or evidence in detail.
0 engagement1 sourcerssFri, Oct 2, 2026, 02:31 AM
Social·Community
AlgoMart Builds AI Harness with Jev, Reserving LLM Calls for Decisions That Need Them
AlgoMart argues fast non-LLM decisions fit inside agentic systems in 'Building an AI Harness with Jev'
AlgoMart argues fast non-LLM decisions fit inside agentic systems in 'Building an AI Harness with Jev'
0 engagement1 sourcerssThu, Sep 24, 2026, 04:33 AM
Products·launched
notjev – turn any OpenAI-compatible LLM into a Jev-style decision engine via single-token logprobs
notjev is a Node library that makes any OpenAI-compatible model produce fast multiple-choice decisions. It requests one token with logprobs enabled, renormalizes the letter probabilities, and returns a verdict, probability, margin, and abstention threshold. This gives Jev-style Choice/Score/Noul outputs to developers running local models (vLLM, llama.cpp, Ollama) without runtime dependencies.
notjev is a Node library that makes any OpenAI-compatible model produce fast multiple-choice decisions. It requests one token with logprobs enabled, renormalizes the letter probabilities, and returns a verdict, probability, margin, and abstention threshold. This gives Jev-style Choice/Score/Noul outputs to developers running local models (vLLM, llama.cpp, Ollama) without runtime dependencies.
29 engagement1 sourceredditTue, Sep 22, 2026, 08:57 AM