phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-24 10:14:26 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fix: price cached_tokens when using custom handler

Browse source
This commit is contained in:
yeahyung 2025-03-10 16:02:32 +09:00
parent 995fc001e0
commit 24101fc7af
1 changed files with 5 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

6
litellm/cost_calculator.py
View file

@ -291,7 +291,7 @@ def cost_per_token( # noqa: PLR0915
if model_info["input_cost_per_token"] > 0:
## COST PER TOKEN ##
prompt_tokens_cost_usd_dollar = (
model_info["input_cost_per_token"] * prompt_tokens
model_info["input_cost_per_token"] * (prompt_tokens - cache_read_input_tokens)
)
elif (
model_info.get("input_cost_per_second", None) is not None
@ -307,6 +307,10 @@ def cost_per_token( # noqa: PLR0915
prompt_tokens_cost_usd_dollar = (
model_info["input_cost_per_second"] * response_time_ms / 1000 # type: ignore
)
## Prompt Caching cost calculation
if model_info["cache_read_input_token_cost"] is not None and model_info["cache_read_input_token_cost"] > 0:
## COST PER TOKEN ##
prompt_tokens_cost_usd_dollar += model_info["cache_read_input_token_cost"] * cache_read_input_tokens
if model_info["output_cost_per_token"] > 0:
completion_tokens_cost_usd_dollar = (