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 Turkish Benchmark Results
A developer shares Jev's performance on Turkish tasks: 44/100 on TRLawBench-3 and 59.42% on Turkish MMLU Pro, below gpt-6-astra's 81.38% and gemma4:31b's 63.64%.
Jev'in Türkçe içerikteki yetenekleri soruluyor. Halihazırda TRLawBench-3'te test etmiştim 44/100 yaptı. Bunun haricinde magibu ekibi de kendi testlerini yapmış. Turkish MMLU Pro'da test etmişler sonuçlar aşağıda: gpt-6-astra (API): %81,38 gemma4:31b: %63,64 Jev: %59,42
· 1 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.
