handle flaky pytests

.circleci/config.yml

@@ -31,6 +31,7 @@ jobs:
             python -m pip install --upgrade pip
             python -m pip install -r .circleci/requirements.txt
             pip install "pytest==7.3.1"
+            pip install "pytest-retry==1.6.3"
             pip install "pytest-asyncio==0.21.1"
             pip install mypy
             pip install "google-generativeai==0.3.2"
@@ -171,6 +172,7 @@ jobs:
             python -m pip install --upgrade pip
             python -m pip install -r .circleci/requirements.txt
             pip install "pytest==7.3.1"
+            pip install "pytest-retry==1.6.3"
             pip install "pytest-mock==3.12.0"
             pip install "pytest-asyncio==0.21.1"
             pip install mypy
@@ -282,6 +284,7 @@ jobs:
           name: Install Dependencies
           command: |
             pip install "pytest==7.3.1"
+            pip install "pytest-retry==1.6.3"
             pip install "pytest-asyncio==0.21.1"
             pip install aiohttp
             pip install "openai==1.40.0"

litellm/tests/test_lowest_latency_routing.py

@@ -299,6 +299,7 @@ async def _gather_deploy(all_deploys):
 @pytest.mark.parametrize(
     "ans_rpm", [1, 5]
 )  # 1 should produce nothing, 10 should select first
+@pytest.mark.flaky(retries=3, delay=1)
 def test_get_available_endpoints_tpm_rpm_check_async(ans_rpm):
     """
     Pass in list of 2 valid models

litellm/tests/test_openai_batches_and_files.py

@@ -95,6 +95,7 @@ def test_create_batch(provider):
 
 @pytest.mark.parametrize("provider", ["openai", "azure"])
 @pytest.mark.asyncio()
+@pytest.mark.flaky(retries=3, delay=1)
 async def test_async_create_batch(provider):
     """
     1. Create File for Batch completion

litellm/tests/test_router_caching.py

@@ -1,7 +1,11 @@
 #### What this tests ####
 #    This tests caching on the router
-import sys, os, time
+import asyncio
-import traceback, asyncio
+import os
+import sys
+import time
+import traceback
+
 import pytest
 
 sys.path.insert(
@@ -71,6 +75,7 @@ def test_router_sync_caching_with_ssl_url():
 
 
 @pytest.mark.asyncio
+@pytest.mark.flaky(retries=3, delay=1)
 async def test_acompletion_caching_on_router():
     # tests acompletion + caching on router
     try:

tests/test_openai_endpoints.py

@@ -325,6 +325,7 @@ async def test_chat_completion_ratelimit():
 
 
 @pytest.mark.asyncio
+@pytest.mark.flaky(retries=3, delay=1)
 async def test_chat_completion_different_deployments():
     """
     - call model group with 2 deployments