litellm-mirror/litellm/caching/_internal_lru_cache.py
Ishaan Jaff d88f01d518
(litellm SDK perf improvements) - handle cases when unable to lookup model in model cost map (#7750)
* use lru cache wrapper

* use lru_cache_wrapper for _cached_get_model_info_helper

* fix _get_traceback_str_for_error

* huggingface/mistralai/Mistral-7B-Instruct-v0.3
2025-01-13 19:58:46 -08:00

30 lines
794 B
Python

from functools import lru_cache
from typing import Callable, Optional, TypeVar
T = TypeVar("T")
def lru_cache_wrapper(
maxsize: Optional[int] = None,
) -> Callable[[Callable[..., T]], Callable[..., T]]:
"""
Wrapper for lru_cache that caches success and exceptions
"""
def decorator(f: Callable[..., T]) -> Callable[..., T]:
@lru_cache(maxsize=maxsize)
def wrapper(*args, **kwargs):
try:
return ("success", f(*args, **kwargs))
except Exception as e:
return ("error", e)
def wrapped(*args, **kwargs):
result = wrapper(*args, **kwargs)
if result[0] == "error":
raise result[1]
return result[1]
return wrapped
return decorator