phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 18:54:30 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix(utils.py): support get_max_tokens() call with same model_name as completion

Browse source 
Closes https://github.com/BerriAI/litellm/issues/3921
This commit is contained in:
Krrish Dholakia 2024-05-31 21:37:25 -07:00
parent b8df5d1a01
commit 7523f803d2
2 changed files with 19 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

14
litellm/tests/test_utils.py
View file

@ -22,6 +22,7 @@ from litellm.utils import (
token_counter,
create_pretrained_tokenizer,
create_tokenizer,
get_max_tokens,
)
# Assuming your trim_messages, shorten_message_to_fit_limit, and get_token_count functions are all in a module named 'message_utils'
@ -372,3 +373,16 @@ def test_supports_function_calling():
assert litellm.supports_function_calling(model="claude-2") == False
except Exception as e:
pytest.fail(f"Error occurred: {e}")
def test_get_max_token_unit_test():
"""
More complete testing in `test_completion_cost.py`
"""
model = "bedrock/anthropic.claude-3-haiku-20240307-v1:0"
max_tokens = get_max_tokens(
model
) # Returns a number instead of throwing an Exception
assert isinstance(max_tokens, int)