Skip to content

[Models] Prevent CUDA sync in Qwen2.5-VL #24741

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Sep 12, 2025
+4 −1
Merged

[Models] Prevent CUDA sync in Qwen2.5-VL #24741

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
bb2bcf3
[Models] Prevent CUDA sync in Qwen2.5-VL
lgeiger Sep 12, 2025
6787c54
Merge branch 'main' into qwen-gpu-sync
lgeiger Sep 12, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion vllm/model_executor/models/qwen2_5_vl.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,6 +64,7 @@
from vllm.platforms import _Backend
from vllm.sequence import IntermediateTensors
from vllm.transformers_utils.config import uses_mrope
from vllm.utils import is_pin_memory_available
from vllm.utils.tensor_schema import TensorSchema, TensorShape

from .interfaces import (MultiModalEmbeddings, SupportsLoRA,
Expand Down Expand Up @@ -737,7 +738,7 @@ def compute_attn_mask_seqlen(
@staticmethod
def invert_permutation(perm: torch.Tensor) -> torch.Tensor:
# building the inverse permutation in O(n) time
inv = torch.empty_like(perm)
inv = torch.empty_like(perm, pin_memory=is_pin_memory_available())
inv[perm] = torch.arange(perm.numel(),
device=perm.device,
dtype=perm.dtype)
Expand Down Expand Up @@ -808,6 +809,8 @@ def forward(
non_blocking=True)
window_index = window_index.to(device=hidden_states.device,
non_blocking=True)
reverse_indices = reverse_indices.to(device=hidden_states.device,
non_blocking=True)

hidden_states = hidden_states.reshape(
seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1)
Expand Down