Automated README summary
Automatically summarized from the README on 1 Nov 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A benchmark for System One decision models that read a state and answer typed probabilistic questions. It scores 20 models on frozen, byte-identical requests across 16 text datasets (English, French, Arabic) plus a 6-dataset vision track, and publishes a leaderboard.
How it uses Jev
Jev 1.13 and preview Jev 1.14 are among the models benchmarked. The harness sends each model identical state plus questions bodies (noul, choice, score) via the OpenRouter Decisions API or the TypeSafe API, then scores the returned probabilities for accuracy, calibration, latency and cost.
Primitives:choicescorenoul
Technique worth stealing
Freeze the data and send byte-identical state and questions to every model, then score accuracy, calibration, latency and cost with one shared scorer.
Try it
Clone the repo, build the suites, set OPENROUTER_API_KEY, run run_decisions.py --model typesafe/jev-1.14, then score.py.
Model judgments from README excerpts
Project signals use repository metadata, an existing summary and the first 700 README characters, not the full project. “Not found” means not detected in that material; it does not establish absence. These are the original judgments with recorded models and dates, not source audits or run results.
Models & evaluation dates
- Inclusion gate
jev-1.14.0- Project signals
jev-1.14.0- Category
jev-1.14.0- Pattern tags
jev-1.14.0
Evidence
These signals use repository metadata, an existing summary and at most 700 cleaned README characters. ≥ 0.60 means reported in that excerpt; ≤ 0.40 means not found there. The middle band is inconclusive. None is a source-code audit or a test result. The runnable-tool question excludes research experiments; the recommendation question asks for a learning shortlist, not a quality verdict.
- Jev is core to the projectnot found in the excerpt0.36
- Shows a System One patternunclear0.44
- Describes handling uncertaintynot found in the excerpt0.02
- Reports measured numbersnot found in the excerpt0.22
- Is a runnable toolnot found in the excerpt0.08
- Top-10 learning shortlist signalnot found in the excerpt0.06
- Is a replica of the modelnot found in the excerpt0.09
- Solves a common problem (0–2)score on a 0–2 scale1.29 / 2
- About JevFound in material0.90
no-judgment band (0.40–0.60)
no-judgment band (0.40–0.60)
The Jev model and date for each judgment stage are listed above. Card written by DeepSeek V4.1 Flash from the README on 1 Nov 2026.