x.comthreadbrrrock (@bchap1n) on X
Announcement of an updated model card for JEV-27B-VL, an open-weight vision-language model released under Apache-2.0. The card covers the model's weights and a "decision adapter," a BoE (Bag/Blend of Experts) architecture paired with a System 1 / System 2 reasoning setup, calibration details and decision prompts, serving instructions for vLLM and SGLang, and an evaluation of image recommendation quality.
TL;DR Open-weight JEV-27B-VL model card update adds decision-adapter, BoE/System1-System2 architecture details, calibration data, and vLLM/SGLang serving instructions.
Organisations named here
- Jev
- SGLang
- vLLM
Technologies named here
- vLLM
- JEV-27B-VL
- SGLang
- Vision language model
Techniques named here
- BoE architecture
- System 1 / System 2 setup
- decision adapter
Related in models
