Skip to content

[Bugfix] Fix handling of encode_video output in vllm.py so each frame’s Base64 #754

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
kcz358 merged 3 commits into from
Jul 12, 2025
Merged

[Bugfix] Fix handling of encode_video output in vllm.py so each frame’s Base64 #754

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e72a64c
Fix handling of encode_video output in vllm.py so each frame’s Base64…
LiamLian0727 Jul 11, 2025
40141aa
Update vllm.py
LiamLian0727 Jul 11, 2025
e1bcb88
Update vllm.py
LiamLian0727 Jul 11, 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
8 changes: 5 additions & 3 deletions lmms_eval/models/vllm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,8 +150,10 @@ def encode_video(self, video_path):
def flatten(self, input):
new_list = []
for i in input:
for j in i:
new_list.append(j)
if isinstance(i, (list, tuple)):
new_list.extend(i)
else:
new_list.append(i)
return new_list

def generate_until(self, requests) -> List[str]:
Expand Down Expand Up @@ -203,7 +205,7 @@ def generate_until(self, requests) -> List[str]:
messages = [{"role": "user", "content": []}]
# When there is no image token in the context, append the image to the text
messages[0]["content"].append({"type": "text", "text": contexts})
for img in imgs:
for img in self.flatten(imgs):
messages[0]["content"].append({"type": "image_url", "image_url": {"url": f"data:image/png;base64,{img}"}})

batched_messages.append(messages)
Expand Down
Loading