Kev: Open-source Jev-like decision models based on Qwen3.5 released
A new series of open-source decision models named Kev has been introduced as an alternative to Jev, according to posts on r/LocalLLaMA. These models, based on Qwen3.5, are available in 0.8B, 4B, and 9B parameter sizes and are designed for local training and execution. The developers claim the 9B version can run on a 32GB Mac, with side-by-side tests against Jev also reported by the first-party source.
2 articles from 1 outlet covered this story. The underlying claim is sourced from a shipped.
What do all outlets agree on?
1 outlet covered “Kev: Open-source Jev-like decision models based on Qwen3.5 released”. All of them report the following:
- Kev models are open-source decision models
- Kev models are an alternative to Jev
- Kev models are based on Qwen3.5
- Kev models are available in 0.8B, 4B, and 9B parameter sizes
- Kev models can be trained and run locally
- The 9B Kev model fits a 32GB Mac
Which outlets covered this?
All 2 articles found on this story, grouped by the stance of the piece. Every link goes to the original publisher.
pragmatist2