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 vs Gemini: 300 customer requests benchmark
Across 300 customer requests, asking useful what ifs made both models faster and more accurate. Gemini still won on accuracy, but Jev caught more wrong answers while wrongly rejecting 70% fewer correct ones, with verification costing about 1/11 as much.
6/7 So, across 300 customer requests: Asking the useful “what ifs” together made BOTH models faster and more accurate. Gemini still won on accuracy. Jev caught more wrong answers while wrongly rejecting 70% fewer correct ones. With verification, it cost about 1/11th as much
· 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.
