Automated README summary
Automatically summarized from the README on 22 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A reproducible benchmark that evaluates Jev (TypeSafe System One) as a Chinese RAG reranker on C-MTEB/T2Reranking hard negatives, comparing against BM25 and community rerankers with published logs and cost caps.
How it uses Jev
Jev scores each query-passage pair from a pool of 1 positive and 9 hard negatives, using score-zh (Score) and noul-zh (Noul) variants; the returned probabilities rank passages, and nDCG@5, nDCG@10, MRR, Recall and AUC are computed from the JSONL logs.
Primitives:scorenoul
Technique worth stealing
Paired per-query bootstrap confidence intervals over a frozen protocol with pinned model versions and full per-call logs.
Try it
uv run python scripts/prepare_data.py --n 100 --out s1, then run_bench.py with --variant score-zh or noul-zh, then analyze.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 projectunclear0.47
- Shows a System One patternnot found in the excerpt0.04
- Describes handling uncertaintynot found in the excerpt0.03
- Reports measured numbersFound in material0.98
- Is a runnable toolnot found in the excerpt0.05
- Top-10 learning shortlist signalnot found in the excerpt0.24
- Is a replica of the modelnot found in the excerpt0.09
- Solves a common problem (0–2)score on a 0–2 scale1.10 / 2
- About JevFound in material0.99
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 22 Oct 2026.