Ex-OpenAI researcher's TypeSafe AI launches 'Jev', a model that judges instead of writing
The Decoder · rss · 2026-09-16
TypeSafe AI, founded by ex-OpenAI researcher Diogo Almeida, launches 'Jev': a model that deliberately generates no text, outputting pure classifications for software use cases with response times from 70 ms and extremely low token prices. It doesn't prevent mistakes — it only guarantees staying within preset options.
More from Models
- ChatGPT users frustrated as message editing disappears, forcing new chats — Innomen · 2026-09-17
- Code Arena to Reveal Win Rate of #1 GPT-6 Astra vs #2 Claude Fable 5.1 — arena · 2026-09-17
- Stealth Model Union Alpha Free for a Week, Built for Agentic Coding — AiBreakfast · 2026-09-17
- Open-ended RL is 'orbit by freefall': only progress per GPU-time matters — tensorqt · 2026-09-17
- LLMs' subtler hallucination: confidently winging plans for tasks they know nothing about — generativist · 2026-09-17
- Users note Claude's 5th-gen models obsessively spotlight corrections over substance — repligate · 2026-09-17