biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Model] Support Qwen3-VL Model Series (#24727)

Browse Source 
Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Roger Wang
2025-09-16 22:01:04 -07:00
committed by GitHub
parent 5801e49776
commit 0f7acdd73c
13 changed files with 2084 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/models/qwen2.py
View File

@@ -285,7 +285,7 @@ class Qwen2Model(nn.Module):
decoder_layer_type: type[nn.Module] = Qwen2DecoderLayer):
super().__init__()
config = vllm_config.model_config.hf_config
config = vllm_config.model_config.hf_config.get_text_config()
cache_config = vllm_config.cache_config
quant_config = vllm_config.quant_config