Automated README summary
Automatically summarized from the README on 18 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
An open benchmark for developers and evaluators comparing 19 guardrail systems — decision models, hosted Decisions APIs and Bedrock Guardrails — on the same 9,975 labelled checks, scored on accuracy and cost.
How it uses Jev
Jev 1.13.0 and 1.14.0 are run as guardrail systems: each labelled row is sent as a yes/no question, Jev returns a probability, and 0.5 or more counts as yes. The scorer then computes balanced accuracy with paired bootstrap intervals and Holm-adjusted tiers, the same way for every system.
Primitives:choice
Technique worth stealing
Score every system identically: a probability threshold of 0.5 for yes, balanced accuracy, paired bootstrap intervals and Holm-adjusted tiers, with no per-model tuning.
Try it
uv sync --extra dev; run pytest; copy .env.example to .env and add API keys; run scripts in benchmark/runs/.
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 projectnot found in the excerpt0.05
- Shows a System One patternnot found in the excerpt0.26
- Describes handling uncertaintynot found in the excerpt0.05
- Reports measured numbersnot found in the excerpt0.39
- Is a runnable toolnot found in the excerpt0.08
- Top-10 learning shortlist signalnot found in the excerpt0.19
- Is a replica of the modelnot found in the excerpt0.05
- Solves a common problem (0–2)score on a 0–2 scale1.91 / 2
- About JevFound in material0.93
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 18 Oct 2026.