Jev introduces a new shape of LLM - System One, aka Decision Models
- ID
- 27032
- Status
- summarized
- Published
- 22 Sep 2026, 7:09 AM
- Fetched
- 23 Sep 2026, 12:47 AM
- Provider
- Simon Willison
- Category
- developer-ai
- Original URL
- https://simonwillison.net/2026/Sep/21/jev/
- Source URL
- https://simonwillison.net/atom/everything/
Summary
- Score
- 7.5
- Created
- 23 Sep 2026, 12:47 AM
- Tags
- Audience
- developersai_ml_learnersai_agent_userssaas_founders
What happened
TypeSafe AI unveiled Jev, a new model category they call 'System One' (or 'decision models') that accepts text input but returns floating-point numbers instead of text—confidence scores for yes/no questions, probability distributions across choices, or numeric scores along a defined range. Input is priced at $0.042/million tokens with output free, undercutting GPT-5 Nano ($0.05/million), and questions are evaluated in parallel so you can batch many against a single document at similar latency to one.
Why it matters
If you currently use a text-generating LLM for classification, labeling, spam detection, prioritization, or search reranking, Jev's input-only pricing and parallel question evaluation could cut those costs dramatically—especially for high-volume pipelines like BM25 fetch + rerank. Evaluate whether your classification workloads can be expressed as Noul/Choice/Score questions and benchmark Jev against your current approach before committing.
Discussion angle
Where in your existing stack are you paying output-token prices for an LLM that's really just making a classification decision—and could a decision-model API like Jev replace that call at a fraction of the cost?