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

(test) proxy - load test

Browse source
This commit is contained in:
ishaan-jaff 2024-01-03 16:16:09 +05:30
parent 344e8d8508
commit 4680a26e2e
1 changed files with 6 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

9
litellm/proxy/tests/load_test_completion.py
View file

@ -4,16 +4,19 @@ import uuid
import traceback import traceback
litellm_client = AsyncOpenAI(base_url="http://0.0.0.0:8001", api_key="any") litellm_client = AsyncOpenAI(base_url="http://0.0.0.0:8000", api_key="any")
async def litellm_completion(): async def litellm_completion():
# Your existing code for litellm_completion goes here # Your existing code for litellm_completion goes here
try: try:
response = await litellm_client.chat.completions.create( response = await litellm_client.chat.completions.create(
model="azure-gpt-3.5", model="Azure OpenAI GPT-4 Canada-East (External)",
stream=True,
messages=[{"role": "user", "content": f"This is a test: {uuid.uuid4()}"}], messages=[{"role": "user", "content": f"This is a test: {uuid.uuid4()}"}],
) )
async for chunk in response:
print(chunk)
return response return response
except Exception as e: except Exception as e:
@ -26,7 +29,7 @@ async def litellm_completion():
async def main(): async def main():
for i in range(1000000): for i in range(1000000):
start = time.time() start = time.time()
n = 500 # Number of concurrent tasks n = 1000 # Number of concurrent tasks
tasks = [litellm_completion() for _ in range(n)] tasks = [litellm_completion() for _ in range(n)]
chat_completions = await asyncio.gather(*tasks) chat_completions = await asyncio.gather(*tasks)