mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-11 03:38:38 +00:00
6c2ede00ac
* test: remove 130 legacy tests owned by stronger unit proofs Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test: list router _embedding and _aembedding as covered via public embedding calls Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
106 lines
3.3 KiB
Python
106 lines
3.3 KiB
Python
#### What this tests ####
|
|
# This tests calling batch_completions by running 100 messages together
|
|
|
|
import asyncio
|
|
import os
|
|
import time
|
|
import pytest
|
|
|
|
import concurrent
|
|
|
|
from dotenv import load_dotenv
|
|
|
|
import litellm
|
|
|
|
from litellm import Router
|
|
|
|
load_dotenv()
|
|
|
|
|
|
# test_multiple_deployments_sync()
|
|
|
|
|
|
# Assuming litellm, router, and executor are defined somewhere in your code
|
|
|
|
|
|
# test_multiple_deployments_parallel()
|
|
@pytest.mark.parametrize("sync_mode", [True, False])
|
|
@pytest.mark.asyncio
|
|
async def test_cooldown_same_model_name(sync_mode):
|
|
# users could have the same model with different api_base
|
|
# example
|
|
# azure/chatgpt, api_base: 1234
|
|
# azure/chatgpt, api_base: 1235
|
|
# if 1234 fails, it should only cooldown 1234 and then try with 1235
|
|
litellm.set_verbose = False
|
|
try:
|
|
print("testing cooldown same model name")
|
|
model_list = [
|
|
{
|
|
"model_name": "gpt-3.5-turbo",
|
|
"litellm_params": {
|
|
"model": "gpt-4.1-nano",
|
|
"api_key": "bad-key",
|
|
"tpm": 90,
|
|
},
|
|
},
|
|
{
|
|
"model_name": "gpt-3.5-turbo",
|
|
"litellm_params": {
|
|
"model": "gpt-4.1-nano",
|
|
"api_key": os.getenv("OPENAI_API_KEY"),
|
|
"tpm": 1,
|
|
},
|
|
},
|
|
]
|
|
|
|
router = Router(
|
|
model_list=model_list,
|
|
redis_host=os.getenv("REDIS_HOST"),
|
|
redis_password=os.getenv("REDIS_PASSWORD"),
|
|
redis_port=int(os.getenv("REDIS_PORT")),
|
|
routing_strategy="simple-shuffle",
|
|
set_verbose=True,
|
|
num_retries=3,
|
|
allowed_fails=0,
|
|
) # type: ignore
|
|
|
|
if sync_mode:
|
|
response = router.completion(
|
|
model="gpt-3.5-turbo",
|
|
messages=[{"role": "user", "content": "hello this request will pass"}],
|
|
)
|
|
print(router.model_list)
|
|
model_ids = []
|
|
for model in router.model_list:
|
|
model_ids.append(model["model_info"]["id"])
|
|
print("\n litellm model ids ", model_ids)
|
|
|
|
# example litellm_model_names ['azure/gpt-4.1-mini-ModelID-64321', 'azure/gpt-4.1-mini-ModelID-63960']
|
|
assert (
|
|
model_ids[0] != model_ids[1]
|
|
) # ensure both models have a uuid added, and they have different names
|
|
|
|
print("\ngot response\n", response)
|
|
else:
|
|
response = await router.acompletion(
|
|
model="gpt-3.5-turbo",
|
|
messages=[{"role": "user", "content": "hello this request will pass"}],
|
|
)
|
|
print(router.model_list)
|
|
model_ids = []
|
|
for model in router.model_list:
|
|
model_ids.append(model["model_info"]["id"])
|
|
print("\n litellm model ids ", model_ids)
|
|
|
|
# example litellm_model_names ['azure/gpt-4.1-mini-ModelID-64321', 'azure/gpt-4.1-mini-ModelID-63960']
|
|
assert (
|
|
model_ids[0] != model_ids[1]
|
|
) # ensure both models have a uuid added, and they have different names
|
|
|
|
print("\ngot response\n", response)
|
|
except Exception as e:
|
|
pytest.fail(f"Got unexpected exception on router! - {e}")
|
|
|
|
|
|
# test_cooldown_same_model_name()
|