phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 10:44:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

(feat) use base_model for azure response_cost

Browse source
This commit is contained in:
ishaan-jaff 2024-02-07 16:33:07 -08:00
parent 705396240e
commit 0764af4392
1 changed files with 29 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

30
litellm/utils.py
View file

@ -1079,10 +1079,20 @@ class Logging:
call_type=self.call_type,
)
else:
# check if base_model set on azure
base_model = (
self.model_call_details.get("litellm_params", {})
.get("metadata", {})
.get("model_info", {})
.get("base_model", None)
)
# base_model defaults to None if not set on model_info
self.model_call_details[
"response_cost"
] = litellm.completion_cost(
completion_response=result, call_type=self.call_type
completion_response=result,
call_type=self.call_type,
model=base_model,
)
verbose_logger.debug(
f"Model={self.model}; cost={self.model_call_details['response_cost']}"
@ -1158,10 +1168,19 @@ class Logging:
if self.model_call_details.get("cache_hit", False) == True:
self.model_call_details["response_cost"] = 0.0
else:
# check if base_model set on azure
base_model = (
self.model_call_details.get("litellm_params", {})
.get("metadata", {})
.get("model_info", {})
.get("base_model", None)
)
# base_model defaults to None if not set on model_info
self.model_call_details[
"response_cost"
] = litellm.completion_cost(
completion_response=complete_streaming_response,
model=base_model,
)
verbose_logger.debug(
f"Model={self.model}; cost={self.model_call_details['response_cost']}"
@ -1479,8 +1498,17 @@ class Logging:
if self.model_call_details.get("cache_hit", False) == True:
self.model_call_details["response_cost"] = 0.0
else:
# check if base_model set on azure
base_model = (
self.model_call_details.get("litellm_params", {})
.get("metadata", {})
.get("model_info", {})
.get("base_model", None)
)
# base_model defaults to None if not set on model_info
self.model_call_details["response_cost"] = litellm.completion_cost(
completion_response=complete_streaming_response,
base_model=base_model,
)
verbose_logger.debug(
f"Model={self.model}; cost={self.model_call_details['response_cost']}"