jevbooks

← All projects

blink

thegovind/blink · Homepage

Typed decisions from open models: probabilities, no generated text.

Classification & Routing0.36Runner-up: Evaluation & Benchmarking
Automated README summary

Automatically summarized from the README on 30 Sept 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.

What it is

blink is a self-hosted typed-decision model family for developers who need Choice, Noul, and Score probabilities over offered options from text or JSON state, without any generated text.

How it uses Jev

blink reads each question's prompt in one forward pass and takes the next-token scores of the offered option letters. It returns a probability for every offered option, so it can only pick among the options you list. Offer "none" or "unclear" when those can happen.

Primitives:choicenoulscore

Technique worth stealing

Read the next-token scores of offered option letters in one forward pass instead of generating text.

Try it

Open the Space, or run blink-4b/serve.py and set TYPESAFE_BASE_URL.

View on GitHub

Start using Jev: API key and your first request →

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 projectunclear0.40
  • Shows a System One patternunclear0.43
  • Describes handling uncertaintynot found in the excerpt0.04
  • Reports measured numbersnot found in the excerpt0.06
  • Is a runnable toolFound in material0.77
  • Top-10 learning shortlist signalnot found in the excerpt0.09
  • Is a replica of the modelFound in material0.72
  • Solves a common problem (0–2)score on a 0–2 scale1.12 / 2
  • About JevFound in material0.84
Signals
Jev is core to the project0.40Shows a System One pattern0.43Describes handling uncertainty0.04Reports measured numbers0.06Is a runnable tool0.77Top-10 learning shortlist signal0.09Is a replica of the model0.7200.51Solves a common problem (0–2)1.12 / 2012

no-judgment band (0.40–0.60)

Design patterns
Questionnaire0.2100.51

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 30 Sept 2026.