Modelpublic

Saraswathy/vlm-mix-commonsense-expert-step100

sourceHugging Faceupdated 1mo agoView on Hugging Face

0likes9downloads

Model Card

qwen3vl4b_commonsense

Public evaluation-ready LoRA adapter from EasyR1 GRPO training at global step

1.Load it on top of Qwen/Qwen3-VL-4B-Instruct. The frozen file hashes are recorded in adapter_manifest.json.