Article
CommunityJev Media & Talks
YourStory: Jev returns structured decisions, not text
YourStory's introduction to Jev, the transformer-based model that returns structured decisions with probabilities instead of paragraphs or code, covering the September 15 stealth exit, the $40M DCVC-led seed, and TypeSafe's pricing claims.
- Format
- Article
- Reading time
- 2 min
- Published by
- Community
- Author
- Rohan Thacker
- Last verified
- Oct 2026
Highlights
- TypeSafe says Jev responds in roughly 70 to 500 ms and is priced at $0.042 per million input tokens, or $42 per billion, with output free.
- YourStory reports the San Francisco startup emerged from stealth on September 15 with a $40 million seed round led by DCVC.
- Jev is transformer-based but never generates paragraphs, code or chatbot replies: it returns structured decisions with probabilities that software can use.
- TypeSafe says the model is trained exclusively on synthetic data with a method called RLCD, and has not disclosed its parameter count or full architecture.
- The company compares Jev with frontier LLMs that generate output sequentially, claiming 40 to 200 times faster results on comparable System One tasks.
Watch out
Copy reviewed against the publisher's search snippet captured on 2026-10-17; the full article returned HTTP 403 to the fetcher, so only the snippet text is reflected. Price, latency, training and speed figures are TypeSafe's claims, and the parameter count and architecture are undisclosed.
Questions about this entry
Short answers drawn from the linked project or page.
A transformer-based model that does not generate paragraphs, code or chatbot replies. Instead it evaluates a piece of information and returns a structured decision, along with probabilities that software can use, per the publisher's snippet.
Diogo Almeida, a former OpenAI researcher and co-inventor of reinforcement learning from human feedback, alongside Erik Gafni and Sasha Sheng. The San Francisco startup emerged from stealth on September 15 with a $40 million seed round led by DCVC.
$0.042 per million input tokens, or $42 per billion tokens, while output tokens are free, per TypeSafe's listed pricing. The model is available in early access through a waitlist, the piece notes.
Roughly 70 to 500 milliseconds. The company compares that with frontier LLMs that generate output sequentially and says Jev can be 40 to 200 times faster for comparable System One-shaped tasks.
TypeSafe says the model uses a parallel sampling method with what it calls Reinforcement Learning for Calibrated Decisions, or RLCD, and that it is trained exclusively on synthetic data. It has not publicly disclosed the parameter count or full architecture.
Yes: a follow-up analysis, 'LLM vs decision model: Does your AI need to talk or decide?', which argues decision models, LLMs and rules are complementary and that a typed output simplifies integration, while warning that Jev can still choose the wrong option.
It warns that a probability is not an explanation, a permission, or a guarantee of accuracy, so a structured decision still needs judgment about when and where to trust it in a product.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Hacker News: Simon Willison's overview of Noul, Choice, and Score, and where a decision model fits.
benwerd53 pointsJev introduces a new shape of LLMRead the thread on Hacker NewsHacker News: swyx posts the Latent Space interview with Diogo Almeida.
swyx4 pointsJev: System One Models for Prod, Not God – With Diogo Almeida, CEO, TypeSafe AIRead the thread on Hacker NewsX: Diogo Almeida in five minutes
X: LLMs vs. Jev, clearly explained
More Jev resources like this
SiliconANGLE reports TypeSafe closed $870M at a $7.5B valuation and explains the model's pitch: three request types, yes/no answers, picking from a list, or a score, return structured output and a confidence number instead of text.
Jev Media & TalksArticle#community#press#funding
Community
Andreessen Horowitz announces it is leading the investment in TypeSafe AI, calling Jev the fastest growing model it has seen at 1 trillion tokens generated in 3 days. The firm claims 1/100 to 1/500 frontier-model cost and 100x speed for classification, and says 25% of Fortune 500 enterprises have integrated Jev.
Jev Media & TalksArticle#community#press#funding
Community
Tom Joseph's analysis walks through Jev's typed primitives and finds three claims that survive scrutiny: the parallel sampler, the price and speed, and the RLCD training whose calibration plot is unpublished.
Jev Media & TalksArticle#community#press#explainer
Community
TechCrunch reports TypeSafe AI raised $870 million at a $7.5 billion valuation, led by Andreessen Horowitz with Sequoia and DCVC, weeks after Jev's September 15 release. The startup claims a third of Fortune 500 companies already use the model.
Jev Media & TalksArticle#community#press#funding
Community
Bloomberg Law reports TypeSafe AI raised about $870M at a $7.5B valuation in a round led by Andreessen Horowitz, weeks after the Jev launch video went viral. The startup says a third of Fortune 500 companies use Jev but declined to name any.
Jev Media & TalksArticle#community#press#funding
Community
The Economic Times profiles Diogo Almeida, who left OpenAI in 2024 and spent about two years developing the idea behind TypeSafe AI. It cites WSJ on the roughly 25% Fortune 500 adoption and a $40 million DCVC raise.
Jev Media & TalksArticle#community#press#explainer
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.