Skip to content

fix qwen3vl ci #4536

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

Closed
shaopeng-666 wants to merge 2 commits into from
Closed

fix qwen3vl ci #4536

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b401b6b
fix qwen3vl ci
shaopeng-666 Nov 28, 2025
f1f559e
fix qwen3vl ci
shaopeng-666 Nov 29, 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
2 changes: 1 addition & 1 deletion tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_torchair_correctness.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@ def sampling_config():
def model_name():
return "wemaster/deepseek_mtp_main_random_bf16"


@pytest.skip("exist OOM error")
def mtp_torchair_correctness(
sampling_config: SamplingParams,
model_name: str,
Expand Down
14 changes: 11 additions & 3 deletions vllm_ascend/models/qwen3_vl.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@
from functools import partial
from typing import Callable, Optional

import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
Expand Down Expand Up @@ -143,14 +144,21 @@ def cal_cos_sin(self, rotary_pos_emb):
def forward(
self,
x: torch.Tensor,
grid_thw: list[list[int]],
grid_thw: torch.Tensor | list[list[int]],
) -> torch.Tensor:
hidden_states = x.to(device=self.device, dtype=self.dtype)
hidden_states = self.patch_embed(hidden_states)

pos_embeds = self.fast_pos_embed_interpolate(grid_thw)
if isinstance(grid_thw, list):
grid_thw_list = grid_thw
grid_thw = np.array(grid_thw, dtype=np.int32)
else:
grid_thw_list = grid_thw.tolist()
grid_thw = grid_thw.numpy()

pos_embeds = self.fast_pos_embed_interpolate(grid_thw_list)
hidden_states = hidden_states + pos_embeds
rotary_pos_emb = self.rot_pos_emb(grid_thw)
rotary_pos_emb = self.rot_pos_emb(grid_thw_list)
grid_thw_tensor = torch.tensor(grid_thw,
device=self.device,
dtype=torch.int32)
Comment on lines +152 to 164
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot Nov 28, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The current implementation for handling grid_thw when it is a torch.Tensor is inefficient. It converts the tensor to a NumPy array using .numpy() (which can cause a GPU-to-CPU data transfer) and then converts it back to a tensor using torch.tensor(). This can be optimized by handling the list and torch.Tensor cases separately to avoid unnecessary conversions.

        if isinstance(grid_thw, list):
            grid_thw_list = grid_thw
            grid_thw_tensor = torch.tensor(grid_thw,
                                           device=self.device,
                                           dtype=torch.int32)
        else:
            grid_thw_list = grid_thw.tolist()
            grid_thw_tensor = grid_thw.to(device=self.device, dtype=torch.int32)

        pos_embeds = self.fast_pos_embed_interpolate(grid_thw_list)
        hidden_states = hidden_states + pos_embeds
        rotary_pos_emb = self.rot_pos_emb(grid_thw_list)

Expand Down
Loading