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

[Model] Support Mistral3 in the HF Transformers format (#15505)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
This commit is contained in:
Michael Goin
2025-04-01 07:10:05 -06:00
committed by GitHub
parent f3aca1ee30
commit 51d7c6a2b2
9 changed files with 723 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
vllm/model_executor/models/vision.py
View File

@@ -69,6 +69,9 @@ def get_vision_encoder_info(
if isinstance(vision_config, CLIPVisionConfig):
return CLIPEncoderInfo(vision_config)
if isinstance(vision_config, PixtralVisionConfig):
# Need to sneak in spatial_merge_size for Mistral3
vision_config.spatial_merge_size = getattr(hf_config,
"spatial_merge_size", 1)
return PixtralHFEncoderInfo(vision_config)
if isinstance(vision_config, SiglipVisionConfig):
return SiglipEncoderInfo(vision_config)