Automated README summary
Automatically summarized from the README on 7 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
gut is a Python library that turns judgment calls—like spam detection, ticket classification, or urgency rating—into one line of code. It is built for TypeSafe AI's Jev but runs on any small model, including local NLI, LLMs, Ollama, vLLM, and OpenAI.
How it uses Jev
Jev is the default backend. gut.likely, gut.classify, and gut.rate send typed questions to Jev, which returns probabilities. The result is used directly in code, e.g., if gut.likely(comment, "is spam"): hide(comment). No prompt or parsing.
Primitives:choicescorenoul
Technique worth stealing
Use a small model made for typed questions to get calibrated probabilities without generation or parsing.
Try it
pip install "gutfeel[jev]" then import gut and configure with JevBackend.
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 projectnot found in the excerpt0.29
- Shows a System One patternFound in material0.64
- Describes handling uncertaintynot found in the excerpt0.11
- Reports measured numbersnot found in the excerpt0.06
- Is a runnable toolFound in material0.96
- Top-10 learning shortlist signalnot found in the excerpt0.22
- Is a replica of the modelFound in material0.64
- Solves a common problem (0–2)score on a 0–2 scale1.99 / 2
- About JevFound in material0.97
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 7 Oct 2026.