From 305821902d41a6b3abe12778fb7ce1329297f174 Mon Sep 17 00:00:00 2001
From: Krrish Dholakia <krrishdholakia@gmail.com>
Date: Tue, 5 Nov 2024 23:24:05 +0530
Subject: [PATCH] test: handle anthropic api instability

---
 tests/local_testing/test_prompt_caching.py | 117 ++++++++++-----------
 1 file changed, 56 insertions(+), 61 deletions(-)

diff --git a/tests/local_testing/test_prompt_caching.py b/tests/local_testing/test_prompt_caching.py
index 35d5e2588..c73bda04e 100644
--- a/tests/local_testing/test_prompt_caching.py
+++ b/tests/local_testing/test_prompt_caching.py
@@ -47,70 +47,65 @@ def _usage_format_tests(usage: litellm.Usage):
     ],
 )
 def test_prompt_caching_model(model):
-    for _ in range(2):
-        response = litellm.completion(
-            model=model,
-            messages=[
-                # System Message
-                {
-                    "role": "system",
-                    "content": [
-                        {
-                            "type": "text",
-                            "text": "Here is the full text of a complex legal agreement"
-                            * 400,
-                            "cache_control": {"type": "ephemeral"},
-                        }
-                    ],
-                },
-                # marked for caching with the cache_control parameter, so that this checkpoint can read from the previous cache.
-                {
-                    "role": "user",
-                    "content": [
-                        {
-                            "type": "text",
-                            "text": "What are the key terms and conditions in this agreement?",
-                            "cache_control": {"type": "ephemeral"},
-                        }
-                    ],
-                },
-                {
-                    "role": "assistant",
-                    "content": "Certainly! the key terms and conditions are the following: the contract is 1 year long for $10/mo",
-                },
-                # The final turn is marked with cache-control, for continuing in followups.
-                {
-                    "role": "user",
-                    "content": [
-                        {
-                            "type": "text",
-                            "text": "What are the key terms and conditions in this agreement?",
-                            "cache_control": {"type": "ephemeral"},
-                        }
-                    ],
-                },
-            ],
-            temperature=0.2,
-            max_tokens=10,
-        )
+    try:
+        for _ in range(2):
+            response = litellm.completion(
+                model=model,
+                messages=[
+                    # System Message
+                    {
+                        "role": "system",
+                        "content": [
+                            {
+                                "type": "text",
+                                "text": "Here is the full text of a complex legal agreement"
+                                * 400,
+                                "cache_control": {"type": "ephemeral"},
+                            }
+                        ],
+                    },
+                    # marked for caching with the cache_control parameter, so that this checkpoint can read from the previous cache.
+                    {
+                        "role": "user",
+                        "content": [
+                            {
+                                "type": "text",
+                                "text": "What are the key terms and conditions in this agreement?",
+                                "cache_control": {"type": "ephemeral"},
+                            }
+                        ],
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "Certainly! the key terms and conditions are the following: the contract is 1 year long for $10/mo",
+                    },
+                    # The final turn is marked with cache-control, for continuing in followups.
+                    {
+                        "role": "user",
+                        "content": [
+                            {
+                                "type": "text",
+                                "text": "What are the key terms and conditions in this agreement?",
+                                "cache_control": {"type": "ephemeral"},
+                            }
+                        ],
+                    },
+                ],
+                temperature=0.2,
+                max_tokens=10,
+            )
+
+            _usage_format_tests(response.usage)
+
+        print("response=", response)
+        print("response.usage=", response.usage)
 
         _usage_format_tests(response.usage)
 
-    print("response=", response)
-    print("response.usage=", response.usage)
-
-    _usage_format_tests(response.usage)
-
-    assert "prompt_tokens_details" in response.usage
-    assert response.usage.prompt_tokens_details.cached_tokens > 0
-
-    # assert "cache_read_input_tokens" in response.usage
-    # assert "cache_creation_input_tokens" in response.usage
-
-    # # Assert either a cache entry was created or cache was read - changes depending on the anthropic api ttl
-    # assert (response.usage.cache_read_input_tokens > 0) or (
-    #     response.usage.cache_creation_input_tokens > 0
-    # )
+        assert "prompt_tokens_details" in response.usage
+        assert response.usage.prompt_tokens_details.cached_tokens > 0
+    except litellm.InternalServerError:
+        pass
 
 
 def test_supports_prompt_caching():