xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-24 11:37:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix pre-commit

Browse Source 
Signed-off-by: bk-201 <joy25810@foxmail.com>
This commit is contained in:
bk-201 2025-12-17 06:57:34 +00:00
parent fe104bd63c
commit 1c8e3c4486
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/v1/worker/gpu_model_runner.py
View File

@ -2185,7 +2185,7 @@ class GPUModelRunner(
# pos_info.length may overcount (e.g., special tokens in Qwen-VL). # pos_info.length may overcount (e.g., special tokens in Qwen-VL).
# Fall back to length if is_embed is None. # Fall back to length if is_embed is None.
num_tokens = self.info.get_num_mm_encoder_tokens( # type: ignore[attr-defined] num_tokens = self.info.get_num_mm_encoder_tokens( # type: ignore[attr-defined]
pos_info.get_num_embeds() pos_info.get_num_embeds
) )
prompt_lora_mapping.append(lora_id) prompt_lora_mapping.append(lora_id)
token_lora_mapping.extend([lora_id] * num_tokens) token_lora_mapping.extend([lora_id] * num_tokens)