CommunityJev Media & Talks
An ex-OpenAI researcher deleted language from the LLM
Fireship on Diogo Almeida's System One launch claims: no chat output, and the headline 200× speed and 400× cost multiples.
- Format
- Video
- Published by
- Community
- Author
- Fireship
- Last verified
- Sep 2026
Works forJev for Classification
Used inJev for Software & DevTools
Highlights
- Description frames Jev as a System 1 model that cannot talk or write code.
- Repeats launch claims of 200× faster and 400× cheaper with free output tokens.
- Includes a Mux sponsor read; the technical claims are restated from TypeSafe's launch.
Watch out
News-style overview, not a measured benchmark. Multiples are vendor marketing unless footnoted in the video.
Questions about this entry
Short answers drawn from the linked project or page.
Not to anything you can check. The fetched page carries only the title and YouTube boilerplate, with no transcript, and the entry notes the multiples are restated from the launch rather than footnoted with a benchmark. They are vendor numbers wearing a news-video costume.
The thread does not document that. With no transcript available, nothing here says what appears on screen or whether any run is reproduced live. What the entry captures is only the framing: a System 1 model that cannot talk or write code.
A recap. The technical claims are restated from the vendor launch, and the clip carries a sponsor read. Since the page documents nothing beyond the title, none of the figures can be verified against the video itself, so keep the recap and the benchmark separate.
No. Nothing published here separates typical from worst-case latency, and no task mix is given behind the multiples. Until an independent benchmark runs on your workload, treat the clip as an explanation of the idea rather than a source for pricing or timeouts.
Fireship published it as a news-style video explainer dated 30 September 2026. The entry classifies it as an overview rather than a measured benchmark.
A Mux sponsor read. The entry notes the technical claims are restated from TypeSafe's launch, and that the multiples are vendor marketing unless footnoted in the video.
As a System 1 model that cannot talk or write code. The entry records that framing and the launch claims of 200x faster and 400x cheaper with free output tokens.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
YouTube: Fireship on System One launch claims: no chat output, 200× speed, 400× cost, and sponsor read.
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
More Jev resources like this
Max Jesch, on IBM Developer, walks Jev as a System One model that returns probabilities for classification instead of generating text.
Jev Media & TalksVideo#community#video#explainer
Community
Martin Keen, on IBM Technology, explains Jev as a System One model whose output is a decision rather than generated text.
Jev Media & TalksVideo#community#video#explainer
Community
Tim Lin's visual explainer on System 1 versus System 2, the Choice, Score, and Noul primitives, and how JevScout uses Jev inside a coding agent.
Jev Media & TalksVideo#community#video#explainer
Community
plain. on Jev as typed answers over state, the September 2026 launch, DCVC seed funding, and Diogo Almeida's InstructGPT background.
Jev Media & TalksVideo#community#video#explainer
Community
VikStack lists ten concepts for Jev: non-autoregressive parallel sampling, 70–500 ms claims, and boolean, enum, and rubric answer shapes.
Jev Media & TalksVideo#community#video#explainer
Community
Innovations for Education explains Jev for no-code and vibe coders when if-then rules need judgment calls like spam or toxicity without burning tokens.
Jev Media & TalksVideo#community#video#explainer
Community
From the Jev community
Posts from builders shipping with Jev right now.