devwithjevJev Evaluated on News and Mixed Data
Jev, accessed through the TypeSafe API, was compared with local model Laya on 60 questions. Jev scored 86.7% on AG News and 71.8% on mixed datasets.
View on X time21.3s
AA Li@alilibx๐
Laya (local, M5 GPU) vs Jev (TypeSafe API), same 60 questions:
AG News: Laya 93.3% in 3.7s vs Jev 86.7% in 21.3s
Mixed datasets: Jev 71.8% vs Laya 50%
Local wins on speed, but only inside its training data. Outside it, Jev generalizes better.
Full race ๐ https://t.co/M7yY3rVsdM
On AG News, Laya scored 93.3% in 3.7s, while Jev took 21.3s. The author says local performance was stronger within its training data, while Jev generalized better outside it.
Also filed under Tools & apps
- Search Telegram Channels by Topic
Sanyi built a tool using JEV that lets people search Telegram channels in his community by topic, question, or vibe. Matching channels rise up in seconds.
- Jev Omni MCP Build
eureka-pd/jev-omni-mcp is an MCP build for Jev.
- Jev Verbose Couscous Build
Beliato's `jev-verbose-couscous` is a build made with Jev.
- Build a Twilio IVR with Jev
This project uses Jev with Twilio to build an interactive voice response system.