phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 11:14:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge pull request #5018 from haadirakhangi/main

Browse source 
Qdrant Semantic Caching
This commit is contained in:
Ishaan Jaff 2024-08-21 08:50:43 -07:00 • committed by GitHub
commit a34aeafdb5
5 changed files with 694 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

10
litellm/utils.py
View file

@ -121,7 +121,7 @@ import importlib.metadata
from openai import OpenAIError as OriginalError
from ._logging import verbose_logger
from .caching import RedisCache, RedisSemanticCache, S3Cache
from .caching import RedisCache, RedisSemanticCache, S3Cache, QdrantSemanticCache
from .exceptions import (
APIConnectionError,
APIError,
@ -1164,6 +1164,14 @@ def client(original_function):
cached_result = await litellm.cache.async_get_cache(
*args, **kwargs
)
elif isinstance(litellm.cache.cache, QdrantSemanticCache):
preset_cache_key = litellm.cache.get_cache_key(*args, **kwargs)
kwargs["preset_cache_key"] = (
preset_cache_key # for streaming calls, we need to pass the preset_cache_key
)
cached_result = await litellm.cache.async_get_cache(
*args, **kwargs
)
else: # for s3 caching. [NOT RECOMMENDED IN PROD - this will slow down responses since boto3 is sync]
preset_cache_key = litellm.cache.get_cache_key(*args, **kwargs)
kwargs["preset_cache_key"] = (