The LiteLLm limits in "evaluation" extra are wrong for Python 3.14
https://github.com/googleapis/python-aiplatform/blob/main/setup.py#L196
The limits are litellm<1.86.0,>=1.83.7; extra == "evaluation"
And they only "accidental" - whole python-aiplatform supports 3.14, the first version that "actually" supports python 3.14 is 1.93.0 - and it only accidentally version below 1.86.2 have no Python 3.14 limit set - but they do not really support 3.14.
This causes problems with other packages for Airflow that are bumping the litellm to >=1.93.0 and they are conflicting. This makes actually "[evaluation]" extra do not work for 3.14 in fact.
Can you please bump lite llm to at least version 1.93.0 in the extra?
The LiteLLm limits in "evaluation" extra are wrong for Python 3.14
https://github.com/googleapis/python-aiplatform/blob/main/setup.py#L196
The limits are
litellm<1.86.0,>=1.83.7; extra == "evaluation"And they only "accidental" - whole python-aiplatform supports 3.14, the first version that "actually" supports python 3.14 is 1.93.0 - and it only accidentally version below 1.86.2 have no Python 3.14 limit set - but they do not really support 3.14.
This causes problems with other packages for Airflow that are bumping the litellm to >=1.93.0 and they are conflicting. This makes actually "[evaluation]" extra do not work for 3.14 in fact.
Can you please bump lite llm to at least version 1.93.0 in the extra?