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

[Model] Use context managers for encoder- and LM-only mode (#32605)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-01-20 11:43:38 +08:00
committed by GitHub
parent 6c01ffb897
commit 4753f3bf69
21 changed files with 290 additions and 353 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
vllm/model_executor/models/step_vl.py
View File

@@ -504,7 +504,8 @@ class StepVLForConditionalGeneration(Step3VLForConditionalGeneration):
self.config = config
self.multimodal_config = multimodal_config
self.use_data_parallel = multimodal_config.mm_encoder_tp_mode == "data"
if multimodal_config.get_limit_per_prompt("image"):
with self._mark_tower_model(vllm_config, "image"):
self.vision_model = PerceptionEncoder(
config.vision_config,
get_act_fn(config.vision_config.hidden_act),
@@ -521,15 +522,13 @@ class StepVLForConditionalGeneration(Step3VLForConditionalGeneration):
prefix=maybe_prefix(prefix, "vit_large_projector"),
disable_tp=self.use_data_parallel,
)
else:
self.vision_model = None
self.vit_large_projector = None
self.language_model = init_vllm_registered_model(
vllm_config=vllm_config,
hf_config=config.text_config,
prefix=maybe_prefix(prefix, "language_model"),
)
with self._mark_language_model(vllm_config):
self.language_model = init_vllm_registered_model(
vllm_config=vllm_config,
hf_config=config.text_config,
prefix=maybe_prefix(prefix, "language_model"),
)
self.make_empty_intermediate_tensors = (
self.language_model.make_empty_intermediate_tensors