Skip to content

feat: support gemini youtube video url #1707

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

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

feat: support gemini youtube video url #1707

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
82185b4
feat: support gemini youtube video url
Sep 12, 2025
da11551
update version
Oct 9, 2025
4af0074
update version
Oct 9, 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 models/vertex_ai/manifest.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,4 +32,4 @@ resource:
tool:
enabled: true
type: plugin
version: 0.0.24
version: 0.0.25
42 changes: 37 additions & 5 deletions models/vertex_ai/models/llm/llm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -510,13 +510,15 @@ def _generate(

history = []
system_instruction = ""


video_metadata = config_kwargs.pop("video_metadata", None)

for msg in prompt_messages:

if isinstance(msg, SystemPromptMessage):
system_instruction = msg.content
else:
content = self._format_message_to_glm_content(msg)
content = self._format_message_to_glm_content(msg, video_metadata=video_metadata)

if history and history[-1].role == content.role:

Expand Down Expand Up @@ -737,7 +739,7 @@ def _convert_one_message_to_text(self, message: PromptMessage) -> str:
raise ValueError(f"Got unknown type {message}")
return message_text

def _format_message_to_glm_content(self, message: PromptMessage) -> glm.Content:
def _format_message_to_glm_content(self, message: PromptMessage, video_metadata: dict = None) -> glm.Content:
"""
Format a single message into glm.Content for Google API

Expand All @@ -755,12 +757,42 @@ def _format_message_to_glm_content(self, message: PromptMessage) -> glm.Content:
elif c.type in [
PromptMessageContentType.IMAGE,
PromptMessageContentType.DOCUMENT,
PromptMessageContentType.AUDIO,
PromptMessageContentType.VIDEO
PromptMessageContentType.AUDIO
]:
data = c.base64_data
mime_type = getattr(c, 'mime_type', None)
parts.append(glm.Part.from_data(data=data, mime_type=mime_type))
elif c.type == PromptMessageContentType.VIDEO:
video_meta = {}
if video_metadata:
if video_metadata.get('start_offset', None):
video_meta['startOffset'] = video_metadata.get('start_offset')
if video_metadata.get('end_offset', None):
video_meta['endOffset'] = video_metadata.get('end_offset')
if c.url:
url = c.url
mime_type = getattr(c, 'mime_type', None)
part_dict = {
"fileData": {
"fileUri": url,
"mimeType": mime_type
}
}
if video_meta:
part_dict['videoMetadata'] = video_meta
parts.append(glm.Part.from_dict(part_dict=part_dict))
else:
data = c.base64_data
mime_type = getattr(c, 'mime_type', None)
part_dict = {
"inlineData": {
"data": data,
"mimeType": mime_type
}
}
if video_meta:
part_dict['videoMetadata'] = video_meta
parts.append(glm.Part.from_dict(part_dict=part_dict))
else:
raise ValueError(f"Unsupported content type: {c.type}")
glm_content = glm.Content(role="user", parts=parts)
Expand Down