Ex-OpenAI RLHF builder launches Jev: a chatless decision model with 150ms responses and $42/B input tokens
SucceededMind · x · 2026-09-16
Diogo Almeida, who helped build RLHF at OpenAI, has launched TypeSafeAI and its first model, Jev. The thesis: generating prose is a major bottleneck for agentic workflows, so structured decisions shouldn't require words.
Jev is a pure API "System One" model — ask a question, get back typed answers with exact confidence scores, no generative text at all. That yields:
- 150ms response times, claimed 200x faster than standard models
- $42 per billion input tokens
- Output tokens free forever
The author frames it as a complement to chat models: route logic and structured decisions to Jev so your if-statements get a decision they can actually trust, while chat models handle the parts that genuinely need generated text.
Related event: ChatGPT Co-inventor's Lab TypeSafe Launches Decision Model Jev with RLCD(10 posts)→
More from coding & agent
- Radio launches: a shared chat room where agents from different providers talk directly — rohanpaul_ai · 2026-09-16
- Celesto open-sources disposable full macOS desktops for AI agents on Apple Silicon — aniketmaurya · 2026-09-16
- AI connector value lies in secrets store and personal context, not payments — jeff_weinstein · 2026-09-16
- Stripe's model-run shop bench: 5 of 7 working stores built by Claude — bcherny · 2026-09-16
- New hire ships five projects in three weeks by syncing team context with /hq-sync — jacob_posel · 2026-09-16
- Anthropic previews Model Hardware Standard to let AI agents run lab instruments — ivan_bezdomny · 2026-09-16