AI model

jevlike

2 mentions in 7 days · 2 in 30 days · 2 total · first seen Sep 16, 2026 · last 21h ago

Timeline2 articles

[AINews] Here are 6 Clones of Jev in 2 days

Jev's non-generative decision model launch drew 36M views and six open clones in two days, positioning discriminative models as fast complements to LLMs.

Latent Space's AI News roundup for 9/17-9/18/2026 centers on Jev, a non-generative 'System 1' decision model whose launch video hit 36M views in two days and reached roughly 13% of teams. At least six open reproductions appeared, including Bespoke Nimble (a Qwen3.5-9B LoRA fine-tune improving a curated eval from 66% to 90% versus Jev's 93%), SemIf (4B/35B Qwen3.5 backbones), and Kev-0.5B built on Qwen2.5-0.5B. Practitioners highlighted calibrated-probability use cases in routing, escalation, and browser/computer-use workflows, with Braintrust reporting ~400x lower scoring cost versus prior setups. The roundup also covers Claude Code v2.1.277 adopting the AGENTS.md standard and research showing coding-agent harness design materially shapes benchmark results.

Latent Spaceupdated Sep 19, 10:46 UTC · 16h agofirst Sep 19, 05:48 UTC · 21h agoAI industry 4 sources1

Reverse-engineered Jev-like model

A GitHub project releases a reverse-engineered, Jev-like model, drawing modest attention with 42 points on Hacker News.

A GitHub repository (vinnylarouge/jevlike) publishes a reverse-engineered recreation described as a Jev-like model. The project surfaced on Hacker News, where it received 42 points and 7 comments. The available source text does not specify the base model, architecture, parameter count, or benchmark results.

Hacker News · AISep 16, 18:49 UTC · 3d agoModel release HN 42↑ · 7 comments1

Appears with

ai model Bespoke Nimble 1 ai model Jev 1 ai model Kev-0.5B 1 ai model Laya 1 ai model Qwen2.5-0.5B 1 ai model Qwen3.5-9B 1 ai model SemIf 1 organization GitHub 1 organization Latent Space 1 product AGENTS.md 1 product Claude Code 1 product Cline 1 product jevlike 1 product LangChain 1 vendor Braintrust 1 vendor Vercel 1

Entities are extracted by the model from each article. Watching an entity keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.