Automated README summary
Automatically summarized from the README on 20 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
Jevulator emulates TypeSafe's Jev HTTP contract locally, returning typed calibrated decisions (noul, choice, score) instead of generated text. It targets developers who want to experiment with Jev-style decisions without an API key, using llama-server and Qwen3.6-35B-A3B.
How it uses Jev
Jevulator emulates Jev rather than using the real model. It renders the chat template once, splits a shared prefix from per-question suffixes, warms the cache, then fires one max_new_tokens=1 call per question and reads logprobs for candidate answer tokens. Results are returned as probabilities for noul, choice, and score questions.
Primitives:noulchoicescore
Technique worth stealing
Render the chat template once, split a shared prefix from per-question suffixes, warm the cache, then read logprobs from single-token calls instead of generating text.
Try it
Run python scripts/run_fixture.py fixtures/<file>.json --url http://127.0.0.1:9931
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 material0.96
- Shows a System One patternnot found in the excerpt0.12
- Describes handling uncertaintynot found in the excerpt0.05
- Reports measured numbersnot found in the excerpt0.05
- Is a runnable toolFound in material0.84
- Top-10 learning shortlist signalnot found in the excerpt0.16
- Is a replica of the modelFound in material0.95
- Solves a common problem (0–2)score on a 0–2 scale0.40 / 2
- About JevFound in material0.98
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 20 Oct 2026.