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

[Model] Add support for nvidia/llama-nemotron-rerank-vl-1b-v2 (#35735)

Browse Source 
Signed-off-by: Jakub Zakrzewski <jzakrzewski@nvidia.com>
This commit is contained in:
Jakub Zakrzewski
2026-03-03 01:32:14 +01:00
committed by GitHub
parent 18c29c746b
commit c8b678e53e
9 changed files with 503 additions and 149 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
vllm/model_executor/models/config.py
View File

@@ -664,6 +664,7 @@ MODELS_CONFIG_MAP: dict[str, type[VerifyAndUpdateConfig]] = {
"LlamaBidirectionalForSequenceClassification": LlamaBidirectionalConfig,
"LlamaBidirectionalModel": LlamaBidirectionalConfig,
"LlamaNemotronVLModel": LlamaNemotronVLConfig,
"LlamaNemotronVLForSequenceClassification": LlamaNemotronVLConfig,
"NomicBertModel": NomicBertModelConfig,
"Qwen2ForProcessRewardModel": Qwen2ForProcessRewardModelConfig,
"Qwen2ForRewardModel": Qwen2ForRewardModelConfig,