From 4d6b7699cc6e512ed2368eb7a04f66372a30c6f4 Mon Sep 17 00:00:00 2001
From: Sameer Kankute <sameer@berri.ai>
Date: Fri, 20 Feb 2026 09:16:24 +0530
Subject: [PATCH] Fix sonnet 3.7 tests

---
 .../test_anthropic_completion.py              | 10 +++++-----
 tests/local_testing/test_caching.py           | 20 +++++++++----------
 2 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/tests/llm_translation/test_anthropic_completion.py b/tests/llm_translation/test_anthropic_completion.py
index e240437178..8630ba6561 100644
--- a/tests/llm_translation/test_anthropic_completion.py
+++ b/tests/llm_translation/test_anthropic_completion.py
@@ -489,7 +489,7 @@ class TestAnthropicCompletion(BaseLLMChatTest, BaseAnthropicChatTest):
 
     def get_base_completion_call_args_with_thinking(self) -> dict:
         return {
-            "model": "anthropic/claude-3-7-sonnet-latest",
+            "model": "anthropic/claude-sonnet-4-5-20250929",
             "thinking": {"type": "enabled", "budget_tokens": 16000},
         }
 
@@ -701,7 +701,7 @@ def test_anthropic_tool_with_image():
     ]
 
     result = prompt_factory(
-        model="claude-3-5-sonnet-20240620",
+        model="claude-sonnet-4-5-20250929",
         messages=messages,
         custom_llm_provider="anthropic",
     )
@@ -761,7 +761,7 @@ def test_anthropic_map_openai_params_tools_and_json_schema():
     mapped_params = litellm.AnthropicConfig().map_openai_params(
         non_default_params=args["non_default_params"],
         optional_params={},
-        model="claude-3-5-sonnet-20240620",
+        model="claude-sonnet-4-5-20250929",
         drop_params=False,
     )
 
@@ -803,7 +803,7 @@ def test_anthropic_map_openai_params_tools_with_defs():
     mapped_params = litellm.AnthropicConfig().map_openai_params(
         non_default_params=args["non_default_params"],
         optional_params={},
-        model="claude-3-5-sonnet-20240620",
+        model="claude-sonnet-4-5-20250929",
         drop_params=False,
     )
 
@@ -1133,7 +1133,7 @@ def test_anthropic_custom_headers():
     with patch.object(client, "post") as mock_post:
         try:
             resp = completion(
-                model="claude-3-5-sonnet-20240620",
+                model="claude-sonnet-4-5-20250929",
                 headers={"anthropic-beta": "computer-use-2025-01-24"},
                 messages=[
                     {"role": "user", "content": "What is the capital of France?"}
diff --git a/tests/local_testing/test_caching.py b/tests/local_testing/test_caching.py
index 7fb57cef9e..3c421e1509 100644
--- a/tests/local_testing/test_caching.py
+++ b/tests/local_testing/test_caching.py
@@ -2647,13 +2647,13 @@ def test_caching_with_reasoning_content():
         litellm.cache = Cache()
 
         response_1 = completion(
-            model="anthropic/claude-3-7-sonnet-latest",
+            model="anthropic/claude-sonnet-4-5-20250929",
             messages=messages,
             thinking={"type": "enabled", "budget_tokens": 1024},
         )
 
         response_2 = completion(
-            model="anthropic/claude-3-7-sonnet-latest",
+            model="anthropic/claude-sonnet-4-5-20250929",
             messages=messages,
             thinking={"type": "enabled", "budget_tokens": 1024},
         )
@@ -2671,14 +2671,14 @@ def test_caching_reasoning_args_miss():  # test in memory cache
         litellm.set_verbose = True
         litellm.cache = Cache()
         response1 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             reasoning_effort="low",
             mock_response="My response",
         )
         response2 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             mock_response="My response",
@@ -2697,14 +2697,14 @@ def test_caching_reasoning_args_hit():  # test in memory cache
         litellm.set_verbose = True
         litellm.cache = Cache()
         response1 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             reasoning_effort="low",
             mock_response="My response",
         )
         response2 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             reasoning_effort="low",
@@ -2724,14 +2724,14 @@ def test_caching_thinking_args_miss():  # test in memory cache
         litellm.set_verbose = True
         litellm.cache = Cache()
         response1 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             thinking={"type": "enabled", "budget_tokens": 1024},
             mock_response="My response",
         )
         response2 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             mock_response="My response",
@@ -2750,14 +2750,14 @@ def test_caching_thinking_args_hit():  # test in memory cache
         litellm.set_verbose = True
         litellm.cache = Cache()
         response1 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             thinking={"type": "enabled", "budget_tokens": 1024},
             mock_response="My response",
         )
         response2 = completion(
-            model="claude-3-7-sonnet-latest",
+            model="claude-4-sonnet-20250514",
             messages=messages,
             caching=True,
             thinking={"type": "enabled", "budget_tokens": 1024},