jev-lab
Awesome-style map of Jev resources plus independent, reproducible benchmarks of primitives, confidence gating and fan-out latency.
Overview
Awesome Jev Lab (llt22) combines a curated guide to the Jev ecosystem with independent tests, raw artifacts and reproduction scripts. The guide points to official resources (docs, models and pricing, patterns such as fan-out, confidence-gated routing, composite scoring and intent routing, the use-case map and the official GitHub organisation), SDKs and integrations, and curated use cases, and warns about similarly named unrelated organisations.
The lab side, per the list summary, benchmarks the Noul, Choice and Score primitives, confidence gating, fan-out latency and agent control, with a validation report dated 18 September 2026 that checks claims against measurements, and a research archive. It is an independent community project available in English and Chinese.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Do Jev's claimed behaviours (primitives, gating, fan-out) hold up when measured?(inferred) | — |
How it works
- Read the ecosystem guide
- Run the lab scripts against Jev
- Compare with the validation report
- Browse the research archive
Strengths
- Guide and measurements in one place
- Raw artifacts for reproduction
- Bilingual
Caveats
- Independent; not official
- Results tied to the tested model version
Getting started
See the 'Run the lab' section of the README.At a glance
- Type
- Benchmark
- Status
- research
- Runtime
- Benchmark scripts
- Has code
- yes
Use cases and integrations
How curated lists describe it
Hands-on research lab for TypeSafe's Jev (System One model): reproducible benchmarks of Noul/Choice/Score primitives, confidence gating, fan-out latency, agent control — plus a living audit of the Jev ecosystem.
logicrw/awesome-jev-projects
Frequently asked questions
What does it contain?
An ecosystem guide plus reproducible benchmarks of Jev's primitives and patterns.
Is it official?
No, it is an independent community project.
Where are the results?
In the Jev Lab validation report in docs/.
Related projects
pytest-style tool that pins fixtures and expected Jev answers, then fails a model upgrade when answers flip or confidence drops.
ProjectbetaPythonTest harness for any Jev-API decision model: accuracy, calibration and robustness probes for reordering, injection and padding.
ServicebetaPythonExperiments testing whether Jev can spot positive-expected-value dice bets from payout tables, and whether recent wins or losses sway it.
ProjectresearchZero-shot run of Jev 1.13 on all seven LexGLUE legal tasks (23,607 examples) versus GPT-5.6 Luna, open models and fine-tuned BERT.
ProjectresearchPythonPre-specified study running Jev and GPT-4.1 as 300 synthetic survey respondents; question format mattered more than model.
ProjectresearchPythonPre-registered paired audit of Jev on Spanish across 3,200 labelled items, plus a CLI to rerun it on your own data.
CLIresearchPython