feature/ introduce new model - gemini #31
Conversation
Added integration tests and unit tests for the Gemini model.
oryandrive
force-pushed
the
feature/add-support-for-gemini-model
branch
from
7532e4a to
e8dc892
Compare
|
Thank you so much for the contribution. We are going to start some testing on our end. |
|
Hey team, thanks! Please let me know if I can help with anything to push that forward. @pgrayy |
|
waiting for this to be merged 🤓🥳 |
There was a problem hiding this comment.
Sorry for the delayed response here, I've gotten a chance to take a look at this pull request. While testing locally, it looks like the original package https://pypi.org/project/google-generativeai/ is now deprecated in favor of https://pypi.org/project/google-genai/. Can you update the dependency introduced in this PR, and rebase it to the latest SDK changes?
| class GeminiModel(Model): | ||
| """Google Gemini model provider implementation.""" | ||
|
|
||
| EVENT_TYPES = { |
There was a problem hiding this comment.
These are unused, please remove.
| "message_stop", | ||
| } | ||
|
|
||
| OVERFLOW_MESSAGES = { |
There was a problem hiding this comment.
Can you share where these are documented?
|
|
||
| client_args = client_args or {} | ||
| genai.client.configure(**client_args) | ||
| self.model = genai.GenerativeModel(self.config["model_id"]) |
There was a problem hiding this comment.
Should this genai.GenerativeModel be passed in instead of the model_id? Does it make sense for this to be configurable?
| """ | ||
| return self.config | ||
|
|
||
| def _format_request_message_content(self, content: ContentBlock) -> dict[str, Any]: |
There was a problem hiding this comment.
ContentBlock has more types like "ToolUse", "ToolResult", "ReasoningContent", and a few other modes. Should this be updated?
| } | ||
|
|
||
| @override | ||
| def format_chunk(self, event: dict[str, Any]) -> StreamEvent: |
There was a problem hiding this comment.
In case gemini returns a streamed response following OpenAI's response stream, you can extend our OpenAI interface to help simplify the logic in this implementation:
Description
Introduce support for the new Model - Gemini
Related Issues
[Link to related issues using #issue-number format]
Documentation PR
[Link to related associated PR in the agent-docs repo]
Type of Change
Testing
hatch fmt --linterhatch fmt --formatterhatch test --allChecklist
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.