mirror of
https://github.com/BerriAI/litellm.git
synced 2026-08-28 05:25:59 +00:00
6a0d03914c
* test: drop the cwd-relative sys.path.insert calls from the test suite
TQ003 stands at 1,077 across 1,058 files, and 1,015 of them are the same shape:
sys.path.insert(0, os.path.abspath("../..")) and its deeper siblings. The
argument resolves against the working directory rather than the file, so from
the repo root, where every job runs pytest, it inserts the directory two levels
above the checkout. It has never pointed at litellm. The package is installed
into the environment anyway, which is what actually makes the import work, and
what the rule's message has said all along.
Removing them leaves 1,634 imports of sys and os with no remaining reference,
and those go too, except where another test module imports the name back out of
the file. The rest of TQ003 is 62 call sites that resolve against __file__ or a
variable, which are a different question and are left alone.
Collection is identical either way: 45,871 tests and the same 51 pre-existing
collection errors before and after, and ruff reports no new undefined name.
* test: drop the duplicate imports the sys.path sweep exposed to F811
* test(pre-call-utils): restore the os import the new bedrock tests need
300 lines
8.9 KiB
Python
300 lines
8.9 KiB
Python
#### What this tests ####
|
|
# This tests the timeout decorator
|
|
|
|
import os
|
|
import traceback
|
|
|
|
import time
|
|
from litellm._uuid import uuid
|
|
|
|
import httpx
|
|
import openai
|
|
import pytest
|
|
|
|
import litellm
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"model, provider",
|
|
[
|
|
("gpt-3.5-turbo", "openai"),
|
|
("azure/gpt-4.1-mini", "azure"),
|
|
],
|
|
)
|
|
@pytest.mark.parametrize("sync_mode", [True, False])
|
|
@pytest.mark.asyncio
|
|
async def test_httpx_timeout(model, provider, sync_mode):
|
|
"""
|
|
Test if setting httpx.timeout works for completion calls
|
|
"""
|
|
timeout_val = httpx.Timeout(10.0, connect=60.0)
|
|
|
|
messages = [{"role": "user", "content": "Hey, how's it going?"}]
|
|
|
|
if sync_mode:
|
|
response = litellm.completion(
|
|
model=model, messages=messages, timeout=timeout_val
|
|
)
|
|
else:
|
|
response = await litellm.acompletion(
|
|
model=model, messages=messages, timeout=timeout_val
|
|
)
|
|
|
|
print(f"response: {response}")
|
|
|
|
|
|
def test_timeout():
|
|
# this Will Raise a timeout
|
|
litellm.set_verbose = False
|
|
try:
|
|
response = litellm.completion(
|
|
model="gpt-3.5-turbo",
|
|
timeout=0.01,
|
|
messages=[{"role": "user", "content": "hello, write a 20 pg essay"}],
|
|
)
|
|
except openai.APITimeoutError as e:
|
|
print(
|
|
"Passed: Raised correct exception. Got openai.APITimeoutError\nGood Job", e
|
|
)
|
|
print(type(e))
|
|
pass
|
|
except Exception as e:
|
|
pytest.fail(
|
|
f"Did not raise error `openai.APITimeoutError`. Instead raised error type: {type(e)}, Error: {e}"
|
|
)
|
|
|
|
|
|
# test_timeout()
|
|
|
|
|
|
def test_bedrock_timeout():
|
|
# this Will Raise a timeout
|
|
litellm.set_verbose = True
|
|
try:
|
|
response = litellm.completion(
|
|
model="bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0",
|
|
timeout=0.01,
|
|
messages=[{"role": "user", "content": "hello, write a 20 pg essay"}],
|
|
)
|
|
pytest.fail("Did not raise error `openai.APITimeoutError`")
|
|
except openai.APITimeoutError as e:
|
|
print(
|
|
"Passed: Raised correct exception. Got openai.APITimeoutError\nGood Job", e
|
|
)
|
|
print(type(e))
|
|
pass
|
|
except Exception as e:
|
|
pytest.fail(
|
|
f"Did not raise error `openai.APITimeoutError`. Instead raised error type: {type(e)}, Error: {e}"
|
|
)
|
|
|
|
|
|
def test_hanging_request_azure():
|
|
"""
|
|
Test that a slow Azure request properly raises APITimeoutError via the Router.
|
|
|
|
Uses a mock to simulate a slow HTTP response so the timeout fires reliably,
|
|
rather than racing against real network latency.
|
|
"""
|
|
litellm.set_verbose = True
|
|
import asyncio
|
|
from unittest.mock import AsyncMock, patch
|
|
|
|
try:
|
|
router = litellm.Router(
|
|
model_list=[
|
|
{
|
|
"model_name": "azure-gpt",
|
|
"litellm_params": {
|
|
"model": "azure/gpt-4.1-mini",
|
|
"api_base": os.environ["AZURE_AI_API_BASE"],
|
|
"api_key": os.environ["AZURE_AI_API_KEY"],
|
|
},
|
|
},
|
|
{
|
|
"model_name": "openai-gpt",
|
|
"litellm_params": {"model": "gpt-3.5-turbo"},
|
|
},
|
|
],
|
|
num_retries=0,
|
|
)
|
|
|
|
encoded = litellm.utils.encode(model="gpt-3.5-turbo", text="blue")[0]
|
|
|
|
original_send = httpx.AsyncClient.send
|
|
|
|
async def _slow_send(self, request, *args, **kwargs):
|
|
await asyncio.sleep(5)
|
|
return await original_send(self, request, *args, **kwargs)
|
|
|
|
async def _test():
|
|
with patch.object(httpx.AsyncClient, "send", new=_slow_send):
|
|
response = await router.acompletion(
|
|
model="azure-gpt",
|
|
messages=[
|
|
{
|
|
"role": "user",
|
|
"content": f"what color is red {uuid.uuid4()}",
|
|
}
|
|
],
|
|
logit_bias={encoded: 100},
|
|
timeout=0.01,
|
|
)
|
|
print(response)
|
|
return response
|
|
|
|
response = asyncio.run(_test())
|
|
|
|
if response.choices[0].message.content is not None:
|
|
pytest.fail("Got a response, expected a timeout")
|
|
except openai.APITimeoutError as e:
|
|
print(
|
|
"Passed: Raised correct exception. Got openai.APITimeoutError\nGood Job", e
|
|
)
|
|
print(type(e))
|
|
pass
|
|
except Exception as e:
|
|
pytest.fail(
|
|
f"Did not raise error `openai.APITimeoutError`. Instead raised error type: {type(e)}, Error: {e}"
|
|
)
|
|
|
|
|
|
# test_hanging_request_azure()
|
|
|
|
|
|
def test_hanging_request_openai():
|
|
litellm.set_verbose = True
|
|
try:
|
|
router = litellm.Router(
|
|
model_list=[
|
|
{
|
|
"model_name": "azure-gpt",
|
|
"litellm_params": {
|
|
"model": "azure/gpt-4.1-mini",
|
|
"api_base": os.environ["AZURE_AI_API_BASE"],
|
|
"api_key": os.environ["AZURE_AI_API_KEY"],
|
|
},
|
|
},
|
|
{
|
|
"model_name": "openai-gpt",
|
|
"litellm_params": {"model": "gpt-3.5-turbo"},
|
|
},
|
|
],
|
|
num_retries=0,
|
|
)
|
|
|
|
encoded = litellm.utils.encode(model="gpt-3.5-turbo", text="blue")[0]
|
|
response = router.completion(
|
|
model="openai-gpt",
|
|
messages=[{"role": "user", "content": "what color is red"}],
|
|
logit_bias={encoded: 100},
|
|
timeout=0.01,
|
|
)
|
|
print(response)
|
|
|
|
if response.choices[0].message.content is not None:
|
|
pytest.fail("Got a response, expected a timeout")
|
|
except openai.APITimeoutError as e:
|
|
print(
|
|
"Passed: Raised correct exception. Got openai.APITimeoutError\nGood Job", e
|
|
)
|
|
print(type(e))
|
|
pass
|
|
except Exception as e:
|
|
pytest.fail(
|
|
f"Did not raise error `openai.APITimeoutError`. Instead raised error type: {type(e)}, Error: {e}"
|
|
)
|
|
|
|
|
|
# test_hanging_request_openai()
|
|
|
|
# test_timeout()
|
|
|
|
|
|
def test_timeout_streaming():
|
|
# this Will Raise a timeout
|
|
litellm.set_verbose = False
|
|
try:
|
|
response = litellm.completion(
|
|
model="gpt-3.5-turbo",
|
|
messages=[{"role": "user", "content": "hello, write a 20 pg essay"}],
|
|
timeout=0.0001,
|
|
stream=True,
|
|
)
|
|
for chunk in response:
|
|
print(chunk)
|
|
except openai.APITimeoutError as e:
|
|
print(
|
|
"Passed: Raised correct exception. Got openai.APITimeoutError\nGood Job", e
|
|
)
|
|
print(type(e))
|
|
pass
|
|
except Exception as e:
|
|
pytest.fail(
|
|
f"Did not raise error `openai.APITimeoutError`. Instead raised error type: {type(e)}, Error: {e}"
|
|
)
|
|
|
|
|
|
# test_timeout_streaming()
|
|
|
|
|
|
@pytest.mark.skip(reason="local test")
|
|
def test_timeout_ollama():
|
|
# this Will Raise a timeout
|
|
import litellm
|
|
|
|
litellm.set_verbose = True
|
|
try:
|
|
litellm.request_timeout = 0.1
|
|
litellm.set_verbose = True
|
|
response = litellm.completion(
|
|
model="ollama/phi",
|
|
messages=[{"role": "user", "content": "hello, what llm are u"}],
|
|
max_tokens=1,
|
|
api_base="https://test-ollama-endpoint.onrender.com",
|
|
)
|
|
# Add any assertions here to check the response
|
|
litellm.request_timeout = None
|
|
print(response)
|
|
except openai.APITimeoutError as e:
|
|
print("got a timeout error! Passed ! ")
|
|
pass
|
|
|
|
|
|
# test_timeout_ollama()
|
|
|
|
|
|
@pytest.mark.parametrize("streaming", [True, False])
|
|
@pytest.mark.parametrize("sync_mode", [True, False])
|
|
@pytest.mark.asyncio
|
|
async def test_anthropic_timeout(streaming, sync_mode):
|
|
litellm.set_verbose = False
|
|
|
|
try:
|
|
if sync_mode:
|
|
response = litellm.completion(
|
|
model="claude-sonnet-4-5-20250929",
|
|
timeout=0.01,
|
|
messages=[{"role": "user", "content": "hello, write a 20 pg essay"}],
|
|
stream=streaming,
|
|
)
|
|
if isinstance(response, litellm.CustomStreamWrapper):
|
|
for chunk in response:
|
|
pass
|
|
else:
|
|
response = await litellm.acompletion(
|
|
model="claude-sonnet-4-5-20250929",
|
|
timeout=0.01,
|
|
messages=[{"role": "user", "content": "hello, write a 20 pg essay"}],
|
|
stream=streaming,
|
|
)
|
|
if isinstance(response, litellm.CustomStreamWrapper):
|
|
async for chunk in response:
|
|
pass
|
|
pytest.fail("Did not raise error `openai.APITimeoutError`")
|
|
except openai.APITimeoutError as e:
|
|
print(
|
|
"Passed: Raised correct exception. Got openai.APITimeoutError\nGood Job", e
|
|
)
|
|
print(type(e))
|
|
pass
|