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
Open Spark Jev is an open-source, local implementation inspired by TypeSafe's Jev and System One, built for NVIDIA DGX Spark. It serves developers who need fast, structured decisions—routing, urgency, tool-call safety, agent continue/stop/ask—rather than generated text.
How it uses Jev
Jev takes a state plus typed questions (Choice, Score, Noul) and returns structured decisions, probabilities, and confidence. In code, MenuScorer.decide() answers multiple questions from one state; classify_tool_call() returns a policy_action from probabilities. A restricted first-token readout replaces text generation.
Primitives:choicescorenoul
Technique worth stealing
Encode state once, then answer each typed question with a short extra pass, returning calibrated probabilities instead of text.
Try it
pip install -e ".[infer]" && osj serve --pull spark-s1-4b-v6
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 projectunclear0.53
- Shows a System One patternnot found in the excerpt0.25
- Describes handling uncertaintynot found in the excerpt0.06
- Reports measured numbersnot found in the excerpt0.05
- Is a runnable toolFound in material0.63
- Top-10 learning shortlist signalnot found in the excerpt0.18
- Is a replica of the modelFound in material0.95
- Solves a common problem (0–2)score on a 0–2 scale1.69 / 2
- About JevFound in material0.92
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.