From dc4235e7f1dd2d40c879cff52c8616bea4cf7efc Mon Sep 17 00:00:00 2001 From: Krrish Dholakia Date: Thu, 14 Nov 2024 18:56:59 +0530 Subject: [PATCH] test: update tests --- .../test_stream_chunk_builder.py | 31 ++++++++++++------- 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/tests/local_testing/test_stream_chunk_builder.py b/tests/local_testing/test_stream_chunk_builder.py index 5fbdf07b8..a65d440a7 100644 --- a/tests/local_testing/test_stream_chunk_builder.py +++ b/tests/local_testing/test_stream_chunk_builder.py @@ -172,6 +172,8 @@ def test_stream_chunk_builder_litellm_usage_chunks(): """ Checks if stream_chunk_builder is able to correctly rebuild with given metadata from streaming chunks """ + from litellm.types.utils import Usage + messages = [ {"role": "user", "content": "Tell me the funniest joke you know."}, { @@ -183,23 +185,28 @@ def test_stream_chunk_builder_litellm_usage_chunks(): {"role": "user", "content": "\nI am waiting...\n\n...\n"}, ] # make a regular gemini call - response = completion( - model="gemini/gemini-1.5-flash", - messages=messages, - ) - usage: litellm.Usage = response.usage + usage: litellm.Usage = Usage( + completion_tokens=64, + prompt_tokens=55, + total_tokens=119, + completion_tokens_details=None, + prompt_tokens_details=None, + ) gemini_pt = usage.prompt_tokens # make a streaming gemini call - response = completion( - model="gemini/gemini-1.5-flash", - messages=messages, - stream=True, - complete_response=True, - stream_options={"include_usage": True}, - ) + try: + response = completion( + model="gemini/gemini-1.5-flash", + messages=messages, + stream=True, + complete_response=True, + stream_options={"include_usage": True}, + ) + except litellm.ServiceUnavailableError as e: + pytest.skip(f"ServiceUnavailableError - {str(e)}") usage: litellm.Usage = response.usage