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

fix(caching.py): support s-maxage param for cache controls

Browse source
This commit is contained in:
Krrish Dholakia 2024-01-04 11:41:23 +05:30
parent 4946b1ef6d
commit b0827a87b2
3 changed files with 13 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

8
litellm/tests/test_caching.py
View file

@ -91,7 +91,7 @@ def test_caching_with_cache_controls():
model="gpt-3.5-turbo", messages=messages, cache={"ttl": 0}
)
response2 = completion(
model="gpt-3.5-turbo", messages=messages, cache={"s-max-age": 10}
model="gpt-3.5-turbo", messages=messages, cache={"s-maxage": 10}
)
print(f"response1: {response1}")
print(f"response2: {response2}")
@ -105,7 +105,7 @@ def test_caching_with_cache_controls():
model="gpt-3.5-turbo", messages=messages, cache={"ttl": 5}
)
response2 = completion(
model="gpt-3.5-turbo", messages=messages, cache={"s-max-age": 5}
model="gpt-3.5-turbo", messages=messages, cache={"s-maxage": 5}
)
print(f"response1: {response1}")
print(f"response2: {response2}")
@ -167,6 +167,8 @@ small text
def test_embedding_caching():
import time
# litellm.set_verbose = True
litellm.cache = Cache()
text_to_embed = [embedding_large_text]
start_time = time.time()
@ -182,7 +184,7 @@ def test_embedding_caching():
model="text-embedding-ada-002", input=text_to_embed, caching=True
)
end_time = time.time()
print(f"embedding2: {embedding2}")
# print(f"embedding2: {embedding2}")
print(f"Embedding 2 response time: {end_time - start_time} seconds")
litellm.cache = None