Automated README summary
Automatically summarized from the README on 10 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A local, self-contained server for Cloudflare/clef-flash, a 9B multimodal decision model. It takes a state and typed questions, returning probability distributions over allowed options. Not a chat model; no text generation.
How it uses Jev
Jev makes decisions by running one forward pass over the Qwen backbone, pooling hidden states for each question and option span, then emitting logits per option. Softmax gives probabilities; Python derives choice, score, confidence, and noul fields.
Primitives:choicescorenoul
Technique worth stealing
Encode state and schema into one token sequence, record token spans for questions/options, and use span pooling with evidence routing to produce per-option logits in a single forward pass.
Try it
make run (downloads weights on first run), then curl http://127.0.0.1:8000/health; make test for one request per question type.
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 projectFound in material0.93
- Shows a System One patternunclear0.52
- Describes handling uncertaintynot found in the excerpt0.04
- Reports measured numbersFound in material0.87
- Is a runnable toolFound in material0.96
- Top-10 learning shortlist signalnot found in the excerpt0.36
- Is a replica of the modelFound in material0.64
- Solves a common problem (0–2)score on a 0–2 scale0.99 / 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 10 Oct 2026.