Ollama
Ollama blog post: Ollama serves the /v1/systemone decision endpoint locally since 0.35, starting with nimble and tev1 models.
This summary is based on how curated lists describe the project; its README was not available to review.
Overview
This Ollama blog post announces support for Jev-style decision models. According to the list summary, Ollama has served the /v1/systemone endpoint locally since version 0.35, and the first decision models in its library are nimble and tev1.
That means developers already running Ollama can send state plus typed questions to a local endpoint and receive per-option probabilities, likely with existing Jev client code pointed at the local server. The post itself was not fetched, so model sizes, accuracy and configuration details are not covered here.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Choice, Score and Noul questions served via /v1/systemone(inferred) | — |
How it works
- Update Ollama to 0.35 or later.
- Pull a decision model such as nimble or tev1.
- Send /v1/systemone requests to the local server.
Strengths
- Local decision models in a widely used runtime.
- No hosted API dependency.
Caveats
- Details beyond the list summary not reviewed.
- Library initially limited to a few decision models.
At a glance
- Type
- Article
- Status
- production
- Runtime
- Ollama
- Has code
- yes
Use cases and integrations
How curated lists describe it
Serves `/v1/systemone` locally since 0.35; the first decision models in its library are `nimble` and `tev1`.
AnotiaWang/awesome-decision-models
Frequently asked questions
Which Ollama version adds this?
Version 0.35, per the list summary.
Which decision models are available?
nimble and tev1 were the first in the library.
Does it need a TypeSafe key?
No, models run locally in Ollama.
Related projects
Docker service exposing local Laya, Von, a CLM head or hosted Jev through one TypeSafe-compatible HTTP and MCP endpoint.
MCP serverbetaPythonConverts open Jev reproductions (Laya, Kev, NanoJev, PlayJev) to INT8 ONNX for offline CPU decisions at about 15.6 ms per question.
ServicebetaPythonNative Go runtime serving Gemma 4 12B as a Jev-like decision model with hand-tuned PTX kernels and no CGo or CUDA toolkit.
ProjectbetaGoGo harness for System One models: hosted Jev, open Kev in-process via llama.cpp, and Cloudflare Clef, with typed helpers and a CLI.
ServicebetaGoGo gateway exposing the Jev /v1/systemone contract over ordinary LLM backends by reading single-token label logprobs.
ServicebetaGoRuns Intern-Decision-0.8B on Apple Silicon behind a local /v1/systemone endpoint that accepts screenshots for agent decisions.
ServicebetaPython