Awesome Jev: early projects and use cases
A community-curated list gathers early Jev projects and ideas for developers who already have API access.
Jev Model Accuracy Comparison: Official Scores Perfect, Variants Vary
A community user compared the accuracy of official Jev, AnyJev (Qwen3-4B), Laya, and djev on a 20-question test. Official Jev answered all correctly, AnyJev scored 95%, Laya 65%, and djev only 35%. djev also changed answers on 19 of 20 questions when option order was shuffled, showing clear instability.
准确率一下就拉开了： 官方 Jev 20 题全对 AnyJev（Qwen3-4B）95% Laya 65% djev（Mac 4bit）35% djev 更离谱，选项换个顺序，20 题里 19 题变了答案（想问问djev，你是属变色龙的吗😂）
· 0 likesOpen on X
More like this
Training an MLP on Qwen 4B to Replicate Jev Behavior
The author shares an open-source project openjev that trains an MLP on qwen 4b to work like Jev, with a Hugging Face link. Link
Swapping an agent's tool-calling reasoning with Jev
The author shares replacing an agent's tool-calling reasoning with Jev, using Fable, Astra, and Opus; early results show drastic savings with very close results, with a demo video promised.
