Automated README summary
Automatically summarized from the README on 29 Sept 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
AutoJev-27B is a multimodal decision model based on Qwen3.8-27B that returns probabilities over supplied choices in one forward pass per question. It ships with a TypeSafe-compatible API, REST server, and browser playground for developers building decision systems.
How it uses Jev
The model itself serves the Jev primitives: POST /v1/systemone accepts choice, noul, and score questions with optional base64 images, returning calibrated probabilities. A DecisionModel loader and server are included. Calibration fits a scalar temperature separately from cross-entropy training.
Primitives:choicenoulscore
Technique worth stealing
Full-weight SFT on 73,000 examples over one H200, with a separate scalar temperature fit for calibration.
Try it
uv sync --frozen --python 3.12, download weights, then AUTOJEV_CHECKPOINT=checkpoints/selected uv run autojev-serve.
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 material0.87
- Shows a System One patternnot found in the excerpt0.15
- Describes handling uncertaintynot found in the excerpt0.04
- Reports measured numbersFound in material0.97
- Is a runnable toolFound in material0.71
- Top-10 learning shortlist signalnot found in the excerpt0.34
- Is a replica of the modelFound in material0.94
- Solves a common problem (0–2)score on a 0–2 scale1.08 / 2
- About JevFound in material0.88
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 29 Sept 2026.