Skip to content

[Bugfix] Fix MRoPE Errors in the Qwen-VL Model When Processing Pure Text #18526

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 3 commits into from
May 22, 2025
Merged

[Bugfix] Fix MRoPE Errors in the Qwen-VL Model When Processing Pure Text #18526

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
ba02e25
[Bugfix] Fix MRoPE Errors in the Qwen-VL Model When Processing Pure Text
May 22, 2025
bd3004e
Update vllm/worker/model_runner.py
DarkLight1337 May 22, 2025
c430224
Merge branch 'main' into fix_qwenvl_pure_text_error_v2
DarkLight1337 May 22, 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
11 changes: 4 additions & 7 deletions vllm/worker/model_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -729,7 +729,10 @@ def _compute_multi_modal_input(self, inter_data: InterDataForSeqGroup,
mm_kwargs, placeholder_maps = MultiModalPlaceholderMap.from_seq_group(
seq_group_metadata,
range(positions[0], positions[0] + len(positions)))
if not mm_kwargs:

# M-RoPE requires mrope_positions even for plain text; return early
# when mm_kwargs is empty only if inter_data.is_prompt is False.
if not mm_kwargs and (not inter_data.is_prompt):
return

inter_data.multi_modal_kwargs = mm_kwargs
Expand All @@ -741,12 +744,6 @@ def _compute_multi_modal_input(self, inter_data: InterDataForSeqGroup,
video_grid_thw = mm_kwargs.get("video_grid_thw", None)
audio_feature_lengths = mm_kwargs.get("audio_feature_lengths",
None)
assert (
image_grid_thw is not None or video_grid_thw is not None
or audio_feature_lengths is not None), (
"mrope embedding type requires multi-modal input mapper "
"returns 'image_grid_thw' or 'video_grid_thw' or "
"'audio_feature_lengths'.")

second_per_grid_ts = mm_kwargs.get("second_per_grid_ts", None)
use_audio_in_video = mm_kwargs.get("use_audio_in_video", False)
Expand Down