User repository on GitHub
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Repos & SDKs
systemone (open typed-decision layer)
An open typed-decision layer with choice, score and noul questions, calibrated probabilities and one batched call, built as a numpy-only base with eight swappable engines. The README positions it as rebuilding Jev's closed API shape on hardware you control.
- Stars
- 1
- Language
- Python
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- Franzferdinan51
- Last verified
- Oct 2026
This is one of several Jev listings for the same project. The canonical entry is Systemone — codesoda.
Used inJev for Software & DevTools
- Languages
- Python
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- Open typed-decision layer: choice, score and noul questions with calibrated probabilities, honest uncertainty, one batched call.
- TypeSafe's Jev is described as a closed API; this package rebuilds that shape on hardware you control, the README says.
- Eight swappable engines: local GLiClass, SGLang, JEV decision models, JevK5, Kev, Cloudflare Clef, a CPU-only ONNX judge and decider-4b.
- DecisionLoop in systemone/loop.py runs See > Decide > Act with cache-friendly prefixes and compressed memory.
- First job: a typed decision about a state — support triage, a guard before a side effect, citation grounding; routing is optional.
Watch out
README and GitHub metadata reviewed 2026-10-17; the digest cuts off inside the DecisionLoop description, so the loop's remaining guarantees, install instructions and API are not captured. No benchmarks or accuracy numbers appear.
Questions about this entry
Short answers drawn from the linked project or page.
An open typed-decision layer from Franzferdinan51 with choice, score and noul questions, calibrated probabilities and honest uncertainty in one batched call, described as rebuilding Jev's shape on hardware you control.
Eight swappable engines: local GLiClass (Apache-2.0, 32M–439M params), SGLang, JEV decision models, JevK5, Kev (0.8B–27B), Cloudflare Clef (9B/27B, multimodal), a CPU-only ONNX judge and the decider-4b sidecar.
Every agentic use runs one loop implemented in systemone/loop.py (DecisionLoop): see the world as text (plus images and video), decide in one batched System 1 call, act and compress the turn into memory.
The README names support triage, a guard before a side effect and citation grounding; picking an effort tier or a model is an optional facet called route.
It is a slim numpy-only base designed to run engines on hardware you control, including local GLiClass and a CPU-only ONNX judge; the README lists hosted-style engines like SGLang too.
The digest shows no accuracy numbers, benchmarks or pricing; the project status and roadmap beyond the three named jobs are not captured in the reviewed material.
MIT, 1 star, Python, created 2026-09-18 and last pushed 2026-10-16, per the digest's GitHub metadata.
More Jev resources like this
A frozen decision server that serves Gemma 4 31B — quantized to FP8 at startup — in TypeSafe's /v1/systemone format, reading the next-token distribution instead of generating tokens. It needs one FP8 NVIDIA GPU with 80 GB and has no authentication.
Jev Repos & SDKs#community#github#python
Community
krishna-gogineni-765/deck31bPython
krishna-gogineni-765/deck31bPythonHTTP server for the Mica v0.5 4B Preview decision model: GGUF weights run through llama.cpp and answer noul, choice and score questions with calibrated probabilities over the TypeSafe /v1/systemone format.
Jev Repos & SDKs#community#github#python
Community
akivet/MicaPython
akivet/MicaPythonServing code for the Jeff 1.0 family of Jev-style decision models: a state, a typed question and its options go in, and one forward pass returns a probability for each option. Four open models from 4B to 31B parameters are served the same way.
Jev Repos & SDKs#community#github#python
Community
j-geuter/jeff-servePython
j-geuter/jeff-servePythonA System One layout optimizer for decision-model workloads: it reorders rows and fields so a prefix-caching backend reuses input computation, then returns per-record decisions in original order. Its bundled backend serves AutoTrust JEV-9B through vLLM.
Jev Repos & SDKs#community#github#python
Community
0814wdwd/solo_jevPython
0814wdwd/solo_jevPythonA FastAPI backend from SiliconLabAI that serves an OpenAI-compatible chat and tool-calling API plus a Jev-style POST /v1/systemone endpoint for typed decisions, powered by cactus-needle on CPU edge devices. The first request downloads the Needle weights once.
Jev Repos & SDKs#community#github#python
Community
siliconlabai/openjev-cactusPython
siliconlabai/openjev-cactusPythonA zero-dependency, copy-anywhere Python service that implements the quick-coarse-verdict-first idea: a deterministic lexicon plus heuristics turn text and structured questions into labels on a controlled vocabulary in micro-to-millisecond time.
Jev Repos & SDKs#community#github#python
Community
junnyofficial/nano_jevPython
junnyofficial/nano_jevPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.