Junhong Liu 59b453eaa2

Speed up mm processor kwargs per request by spliting dynamic and static kwargs (#26483) ...

Signed-off-by: Junhong <liujunhong11@huawei.com>
Signed-off-by: Junhong Liu <98734602+LJH-LBJ@users.noreply.github.com>
Co-authored-by: Junhong <liujunhong11@huawei.com>

2025-11-07 07:51:28 +08:00

..

chat_templates

[Misc] Provide Siglip2 chat template (#27939)

2025-11-02 13:42:38 +00:00

configs

Add llama 4 scaling support (#28145)

2025-11-06 18:55:17 +00:00

processors

[Bugfix] Fix deepseek-ocr multi-image inference and add merge_by_field_config=True with tensor schema support (#27361)

2025-10-22 17:15:38 -07:00

tokenizers

Patch Mistral Tokenizer (#28146)

2025-11-06 06:43:16 +00:00

__init__.py

…

config_parser_base.py

…

config.py

[Chore] Clean up deepseek v2/v3 config copy (#28055)

2025-11-06 03:46:30 +00:00

detokenizer_utils.py

…

dynamic_module.py

…

processor.py

Speed up mm processor kwargs per request by spliting dynamic and static kwargs (#26483)

2025-11-07 07:51:28 +08:00

runai_utils.py

[Chore] Separate out vllm.utils.importlib (#27022)

2025-10-17 00:48:59 +00:00

s3_utils.py

[Chore] Separate out vllm.utils.importlib (#27022)

2025-10-17 00:48:59 +00:00

tokenizer_base.py

…

tokenizer.py

Remove unused imports (#26972)

2025-10-16 19:51:17 -07:00

utils.py

[Bugfix] Skip gs:// model paths for speculator detection (#27846)

2025-11-03 14:31:03 +00:00