Automated README summary
Automatically summarized from the README on 9 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A Python benchmark harness for comparing local System One decision models (julia-1, laya, lev, kev-0.8b, kev-4b, openjev) served by llama.cpp's POST /v1/systemone endpoint, using jev-style metrics on typed-decision suites.
How it uses Jev
Jev-style decision models read a state and answer typed questions (choice, score, noul) with probability distributions in one forward pass. The harness sends these questions to the endpoint, compares returned distributions against gold distributions, and computes calibration and accuracy metrics.
Primitives:choicescorenoul
Technique worth stealing
Matched-weights rule: compare all models at Q8_0 quantization for fair benchmarking.
Try it
uv sync; uv run decision-bench run --suite fixture --mock
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-1.13.0- Project signals
jev-1.13.0- Category
jev-1.13.0- Pattern tags
jev-1.13.0
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 projectunclear0.45
- Shows a System One patternnot found in the excerpt0.12
- Describes handling uncertaintynot found in the excerpt0.04
- Reports measured numbersnot found in the excerpt0.29
- Is a runnable toolnot found in the excerpt0.13
- Top-10 learning shortlist signalnot found in the excerpt0.18
- Is a replica of the modelunclear0.42
- Solves a common problem (0–2)score on a 0–2 scale0.99 / 2
- About JevFound in material0.97
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 9 Oct 2026.