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 async_cache for acompletion/aembedding

Browse source
This commit is contained in:
ishaan-jaff 2023-12-14 16:04:45 +05:30
parent a8e12661c2
commit 008df34ddc
3 changed files with 9 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

4
litellm/utils.py
View file

@ -1682,9 +1682,9 @@ def client(original_function):
# [OPTIONAL] ADD TO CACHE
if litellm.caching or litellm.caching_with_models or litellm.cache != None: # user init a cache object
if isinstance(result, litellm.ModelResponse) or isinstance(result, litellm.EmbeddingResponse):
litellm.cache.add_cache(result.json(), *args, **kwargs)
asyncio.create_task(litellm.cache._async_add_cache(result.json(), *args, **kwargs))
else:
litellm.cache.add_cache(result, *args, **kwargs)
asyncio.create_task(litellm.cache._async_add_cache(result, *args, **kwargs))
# LOG SUCCESS - handle streaming success logging in the _next_ object
print_verbose(f"Async Wrapper: Completed Call, calling async_success_handler: {logging_obj.async_success_handler}")
asyncio.create_task(logging_obj.async_success_handler(result, start_time, end_time))