Jev introduces a new shape of LLM
TypeSafe AI has introduced Jev, a new category of 'System One' or decision models that outputs typed probabilistic decisions rather than text. The model is significantly cheaper and faster than traditional LLMs, charging only for input tokens at a rate lower than OpenAI's GPT-5 Nano. While Jev excels at classification and ranking tasks, its black-box nature raises concerns about hidden biases and the lack of explainability for its outputs.
- ▪Jev accepts text inputs but returns floating-point numbers for categories, yes/no questions, and ratings instead of generating text.
- ▪The service charges only for input tokens at $0.042 per million, making it cheaper than competitors like GPT-5 Nano.
- ▪Users can pose three types of questions to the model: Yes/No (Noul), Choice, and Score questions.
- ▪Jev is particularly effective for classification tasks such as spam detection, label suggestion, and search result reranking.
- ▪The model's opaque decision-making process has sparked debate regarding potential biases and the necessity for rigorous evaluation experiments.
Hacker News (AI / LLM) files mainly under ai. We currently carry 5,922 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Simon Willison’s Weblog |
| Canonical URL | https://simonwillison.net/2026/Sep/21/jev/ |
| Publication time | Tue, 22 Sep 2026 04:38:31 +0000 |
| Retrieval time | 2026-09-22T04:43:51.425Z |
| Last seen | 2026-09-22T04:43:51.425Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | QGvWYz6t3TuH · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Jev introduces a new shape of LLM—System One, aka Decision Models 21st September 2026 Last week TypeSafe AI unveiled Jev, their first example of a new category of model that they are calling “System One models” (I’m with Maggie Appleton, I think “decision models” is a better name for these). Jev is an interesting variant on the usual LLM format: it still accepts text inputs, but instead of text output it returns floating point numbers corresponding to categories, yes/no questions, ratings, and associated confidence scores. TypeSafe describe Jev like this: Think of Jev as a frontier-intelligence function call: unstructured state in, typed probabilistic decisions out. It’s also very fast, and really cheap.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Simon Willison’s Weblog.
Discussion
0 commentsMore from Simon Willison’s Weblog
