Qwen3-VL-8B-Instruct LM head

28f005e about 1 month ago

421 Bytes

library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen3-VL-8B-Instruct

Qwen3-VL-8B-Instruct — LM head

The language-model head (lm_head.weight) of Qwen/Qwen3-VL-8B-Instruct, packaged on its own so a head-less Qwen3-VL feature encoder can be made generative again without loading a second full model — attach this head to the existing encoder body and only this tensor is added to memory.