jevinf

Where it calls Jev

print("  Jev-compatible: POST /v1/systemone  GET /v1/models"

src/jevinf/cli.py:156

The link points at the commit we read, so the line number still holds.

What it asks Jev

These question sets are lifted from this project's source exactly as written.

import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    billing: { type: 'boolean', instructions: 'Is this message about billing?' },
    urgent: { type: 'boolean', instructions: 'Does this need a reply within the hour?' },
    team: { type: 'choice', instructions: 'Which team should handle it?',
      criteria: { payments: 'Payouts, refunds, charges', platform: 'Latency, outages, errors' } },
    tone: { type: 'score', instructions: 'How frustrated is the customer?',
      criteria: ['calm', 'frustrated', 'very angry'] },
  },
});

Other projects in this category

• NanoJev ▶ — A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline.
• kev ▶ — tiny Jev-like model built on top of Qwen2.5-0.5B you can train and run on your MacBook
• reflex ▶ — A small open decision model: state + typed questions -> calibrated probabilities. A Jev / System One re-creation on Qwen3.5.
• von ▶ — The open-source System One decision model. Sub-15ms, non-autoregressive, local drop-in alternative to TypeSafe Jev.
• OpenDecision ▶ — OpenDecision is an open-source semantic decision engine like typesafe's jev.
• jev_local ▶ — Replicating Jev with a local LLM
• OpenJev ▶ — Jev-compatible System 开源Jev
• diffusiongemma-jev-macos ▶ — Local JEV-style decisions with DiffusionGemma on Apple Silicon, with benchmarks and coding-agent examples.