TypeSafe AI Releases Jev: A Decision-Only Model That Returns Typed Probabilities Instead of Text
TypeSafe AI's Jev does what most LLMs won't: return structured probabilities instead of text. Early adopters are embedding it into Vercel and Netlify.
Why it matters
A decision-only model that outputs typed probabilities rather than natural language represents a shift in how AI handles deterministic tasks. Practitioners integrating classification or scoring workflows may find this more efficient than prompt-engineering traditional LLMs for structured output, but the narrow focus and lack of published benchmarks limit its immediate relevance.
The key facts
11 to knowTypeSafe AI founded by former OpenAI researcher Diogo Almeida
Jev outputs typed probabilities and confidence scores instead of text
Evaluates inputs in parallel
Integrations announced with Vercel and Netlify
No benchmark data, pricing, or availability details disclosed
Published October 1, 2026 on InfoQ
Founded by former OpenAI researcher Diogo Almeida
Jev generates typed outputs with probabilistic scores and confidence values
Parallel evaluation of inputs
Integrated into Vercel and Netlify
Positioned as more efficient than traditional (text-generating) models for decision tasks
Go to the source
InfoQ AI/MLinfoq.com
Publisher excerpt: TypeSafe AI, founded by former OpenAI researcher Diogo Almeida, has introduced Jev, a decision-making model that generates typed outputs rather than text. It evaluates inputs in parallel, providing results with probabilistic scores and confidence values. Jev's adoption has been swift, with…