phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

fix(cost_calculator.py): support openai+azure tts calls

Browse source
This commit is contained in:
Krrish Dholakia 2024-07-05 20:58:08 -07:00
parent d528b66db0
commit 407639cc7d
6 changed files with 191 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

48
litellm/cost_calculator.py
View file

@ -13,6 +13,7 @@ from litellm.litellm_core_utils.llm_cost_calc.google import (
from litellm.litellm_core_utils.llm_cost_calc.google import (
cost_per_token as google_cost_per_token,
)
from litellm.litellm_core_utils.llm_cost_calc.utils import _generic_cost_per_character
from litellm.utils import (
CallTypes,
CostPerToken,
@ -62,6 +63,23 @@ def cost_per_token(
### CUSTOM PRICING ###
custom_cost_per_token: Optional[CostPerToken] = None,
custom_cost_per_second: Optional[float] = None,
### CALL TYPE ###
call_type: Literal[
"embedding",
"aembedding",
"completion",
"acompletion",
"atext_completion",
"text_completion",
"image_generation",
"aimage_generation",
"moderation",
"amoderation",
"atranscription",
"transcription",
"aspeech",
"speech",
] = "completion",
) -> Tuple[float, float]:
"""
Calculates the cost per token for a given model, prompt tokens, and completion tokens.
@ -76,6 +94,7 @@ def cost_per_token(
custom_llm_provider (str): The llm provider to whom the call was made (see init.py for full list)
custom_cost_per_token: Optional[CostPerToken]: the cost per input + output token for the llm api call.
custom_cost_per_second: Optional[float]: the cost per second for the llm api call.
call_type: Optional[str]: the call type
Returns:
tuple: A tuple containing the cost in USD dollars for prompt tokens and completion tokens, respectively.
@ -159,6 +178,27 @@ def cost_per_token(
prompt_tokens=prompt_tokens,
completion_tokens=completion_tokens,
)
elif call_type == "speech" or call_type == "aspeech":
prompt_cost, completion_cost = _generic_cost_per_character(
model=model_without_prefix,
custom_llm_provider=custom_llm_provider,
prompt_characters=prompt_characters,
completion_characters=completion_characters,
custom_prompt_cost=None,
custom_completion_cost=0,
)
if prompt_cost is None or completion_cost is None:
raise ValueError(
"cost for tts call is None. prompt_cost={}, completion_cost={}, model={}, custom_llm_provider={}, prompt_characters={}, completion_characters={}".format(
prompt_cost,
completion_cost,
model_without_prefix,
custom_llm_provider,
prompt_characters,
completion_characters,
)
)
return prompt_cost, completion_cost
elif model in model_cost_ref:
print_verbose(f"Success: model={model} in model_cost_map")
print_verbose(
@ -289,7 +329,7 @@ def cost_per_token(
return prompt_tokens_cost_usd_dollar, completion_tokens_cost_usd_dollar
else:
# if model is not in model_prices_and_context_window.json. Raise an exception-let users know
error_str = f"Model not in model_prices_and_context_window.json. You passed model={model}. Register pricing for model - https://docs.litellm.ai/docs/proxy/custom_pricing\n"
error_str = f"Model not in model_prices_and_context_window.json. You passed model={model}, custom_llm_provider={custom_llm_provider}. Register pricing for model - https://docs.litellm.ai/docs/proxy/custom_pricing\n"
raise litellm.exceptions.NotFoundError( # type: ignore
message=error_str,
model=model,
@ -535,6 +575,11 @@ def completion_cost(
raise Exception(
f"Model={image_gen_model_name} not found in completion cost model map"
)
elif (
call_type == CallTypes.speech.value or call_type == CallTypes.aspeech.value
):
prompt_characters = litellm.utils._count_characters(text=prompt)
# Calculate cost based on prompt_tokens, completion_tokens
if (
"togethercomputer" in model
@ -591,6 +636,7 @@ def completion_cost(
custom_cost_per_token=custom_cost_per_token,
prompt_characters=prompt_characters,
completion_characters=completion_characters,
call_type=call_type,
)
_final_cost = prompt_tokens_cost_usd_dollar + completion_tokens_cost_usd_dollar
print_verbose(