Skip to content

[fix] Make Qwen2_5OmniProcessor warning a lot less noisy via warning_once #45455

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
tomaarsen merged 8 commits into from
Apr 16, 2026
+3 −25
Merged

[fix] Make Qwen2_5OmniProcessor warning a lot less noisy via warning_once #45455

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
e20a71f
Make Qwen2_5OmniProcessor warning a lot less noisy via warning_once
tomaarsen Apr 15, 2026
4e0de1b
Bonus request: allow fully disabling the warning via config
tomaarsen Apr 15, 2026
2ad8a4e
Add test to show that warning can be quietened with this
tomaarsen Apr 15, 2026
e3f8ea9
Revert "Make Qwen2_5OmniProcessor warning a lot less noisy via warnin…
tomaarsen Apr 15, 2026
1f77f89
Remove system prompt warning
tomaarsen Apr 15, 2026
9e95eed
Make style
tomaarsen Apr 15, 2026
e0d7086
Fully remove apply_chat_template!
tomaarsen Apr 15, 2026
abc0031
Rerun modular_model_converter.py for qwen3_omni_moe
tomaarsen Apr 15, 2026
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
22 changes: 0 additions & 22 deletions src/transformers/models/qwen2_5_omni/processing_qwen2_5_omni.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,6 @@
Processor class for Qwen2.5Omni.
"""

import logging
import re

import numpy as np
Expand Down Expand Up @@ -301,27 +300,6 @@ def _iter():

return list(_iter())

def apply_chat_template(self, conversations, chat_template=None, **kwargs):
is_batched = False
if isinstance(conversations[0], dict):
conversations = [conversations]
is_batched = True

for conversation in conversations:
if (
conversation[0]["role"] != "system"
or conversation[0]["content"][0]["text"]
!= "You are Qwen, a virtual human developed by the Qwen Team, Alibaba Group, capable of perceiving auditory and visual inputs, as well as generating text and speech."
):
logging.warning(
"System prompt modified, audio output may not work as expected. "
+ "Audio output mode only works when using default system prompt 'You are Qwen, a virtual human developed by the Qwen Team, Alibaba Group, capable of perceiving auditory and visual inputs, as well as generating text and speech.'"
)
Comment on lines -316 to -319
Copy link
Copy Markdown
Member

@zucchini-nlp zucchini-nlp Apr 15, 2026

Choose a reason for hiding this comment

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

Oh I hate this warning as well haha, couldn't convince qwen team to delete at model addition time 🫠

Now that it's maintained by us, ig we can decide what to do with it

if is_batched:
conversations = conversations[0]

return super().apply_chat_template(conversations, chat_template, **kwargs)

def post_process_image_text_to_text(self, generated_outputs, skip_special_tokens=True, **kwargs):
"""
Post-process the output of a vlm to decode the text.
Expand Down
6 changes: 3 additions & 3 deletions src/transformers/models/qwen3_omni_moe/processing_qwen3_omni_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -318,9 +318,6 @@ def _iter():

return list(_iter())

def apply_chat_template(self, conversations, chat_template=None, **kwargs):
return super().apply_chat_template(conversations, chat_template, **kwargs)

def post_process_image_text_to_text(self, generated_outputs, skip_special_tokens=True, **kwargs):
"""
Post-process the output of a vlm to decode the text.
Expand Down Expand Up @@ -392,5 +389,8 @@ def model_input_names(self):
)
)

def apply_chat_template(self, conversations, chat_template=None, **kwargs):
return super().apply_chat_template(conversations, chat_template, **kwargs)


__all__ = ["Qwen3OmniMoeProcessor"]
Loading