Automated README summary
Automatically summarized from the README on 2 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
Benchmarks four architectures for closed-set typed decisions (routing, mapping) on a Jev-style System One interface, measuring accuracy, latency, and cost on Azure AI Foundry and local ONNX.
How it uses Jev
Jev's interface (unstructured state + typed questions in, typed values with probabilities out) is reimplemented across four backends. Each returns a ChoiceAnswer with .choice, .probability, .confidence, .margin; JevClient uses an LLM, the others use encoders.
Primitives:choice
Technique worth stealing
Late interaction: encode state once as a token matrix, condition pooling on the question via per-option attention, halving dilution penalty at bi-encoder cost.
Try it
uv run benchmark_architectures.py; uv run benchmark_scaling.py; uv run benchmark_crossencoder.py --n 200 --seed 111
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 projectunclear0.56
- Shows a System One patternunclear0.40
- Describes handling uncertaintynot found in the excerpt0.07
- Reports measured numbersFound in material0.89
- Is a runnable toolnot found in the excerpt0.14
- Top-10 learning shortlist signalnot found in the excerpt0.33
- Is a replica of the modelFound in material0.89
- Solves a common problem (0–2)score on a 0–2 scale1.90 / 2
- About JevFound in material0.89
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 2 Oct 2026.