fix(ci): switch retired claude-sonnet-4-20250514 to claude-sonnet-4-6 (#975)

tests/e2e/config_memory_state_v1_responses.yaml

@@ -15,7 +15,7 @@ llm_providers:
     default: true
 
     # Anthropic Models
-  - model: anthropic/claude-sonnet-4-20250514
+  - model: anthropic/claude-sonnet-4-6
     access_key: $ANTHROPIC_API_KEY
 
 # State storage configuration for v1/responses API

tests/e2e/test_model_alias_routing.py

@@ -440,7 +440,7 @@ def test_anthropic_thinking_mode_streaming():
     text_delta_seen = False
 
     with client.messages.stream(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         max_tokens=2048,
         thinking={"type": "enabled", "budget_tokens": 1024},  # <- idiomatic
         messages=[{"role": "user", "content": "Explain briefly what 2+2 equals"}],

tests/e2e/test_openai_responses_api_client.py

@@ -489,7 +489,7 @@ def test_openai_responses_api_non_streaming_upstream_anthropic():
     client = openai.OpenAI(api_key="test-key", base_url=f"{base_url}/v1")
 
     resp = client.responses.create(
-        model="claude-sonnet-4-20250514", input="Hello, translate this via grok alias"
+        model="claude-sonnet-4-6", input="Hello, translate this via grok alias"
     )
 
     # Print the response content - handle both responses format and chat completions format
@@ -509,7 +509,7 @@ def test_openai_responses_api_with_streaming_upstream_anthropic():
 
     # Simple streaming responses API request using a direct model (pass-through)
     stream = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="Write a short haiku about coding",
         stream=True,
     )
@@ -566,7 +566,7 @@ def test_openai_responses_api_non_streaming_with_tools_upstream_anthropic():
     ]
 
     resp = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="Call the echo tool",
         tools=tools,
     )
@@ -598,7 +598,7 @@ def test_openai_responses_api_streaming_with_tools_upstream_anthropic():
     ]
 
     stream = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="Call the echo tool with hello_world",
         tools=tools,
         stream=True,

tests/e2e/test_openai_responses_api_client_with_state.py

@@ -35,7 +35,7 @@ def test_conversation_state_management_two_turn():
     # Turn 1: Send initial message to Anthropic (non-OpenAI model)
     logger.info("\n[TURN 1] Sending initial message...")
     resp1 = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="My name is Alice and I like pizza.",
     )
 
@@ -53,7 +53,7 @@ def test_conversation_state_management_two_turn():
         f"\n[TURN 2] Sending follow-up with previous_response_id={response_id_1}"
     )
     resp2 = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="Please list all the messages you have received in our conversation, numbering each one.",
         previous_response_id=response_id_1,
     )
@@ -121,7 +121,7 @@ def test_conversation_state_management_two_turn_streaming():
     # Turn 1: Send initial streaming message to Anthropic (non-OpenAI model)
     logger.info("\n[TURN 1] Sending initial streaming message...")
     stream1 = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="My name is Alice and I like pizza.",
         stream=True,
     )
@@ -154,7 +154,7 @@ def test_conversation_state_management_two_turn_streaming():
         f"\n[TURN 2] Sending follow-up streaming request with previous_response_id={response_id_1}"
     )
     stream2 = client.responses.create(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         input="Please list all the messages you have received in our conversation, numbering each one.",
         previous_response_id=response_id_1,
         stream=True,

tests/e2e/test_prompt_gateway.py

@@ -395,7 +395,7 @@ def test_claude_v1_messages_api():
     )
 
     message = client.messages.create(
-        model="claude-sonnet-4-20250514",  # Use working model from smoke test
+        model="claude-sonnet-4-6",  # Use working model from smoke test
         max_tokens=50,
         messages=[
             {
@@ -414,7 +414,7 @@ def test_claude_v1_messages_api_streaming():
     client = anthropic.Anthropic(api_key="test-key", base_url=base_url)
 
     with client.messages.stream(
-        model="claude-sonnet-4-20250514",
+        model="claude-sonnet-4-6",
         max_tokens=50,
         messages=[
             {
@@ -525,7 +525,7 @@ def test_openai_gpt4o_mini_v1_messages_api_streaming():
 
 
 def test_openai_client_with_claude_model_streaming():
-    """Test OpenAI client using /v1/chat/completions API with Claude model (claude-sonnet-4-20250514)
+    """Test OpenAI client using /v1/chat/completions API with Claude model (claude-sonnet-4-6)
     This tests the transformation: Anthropic upstream -> OpenAI client format with proper chunk handling
     """
     # Get the base URL from the LLM gateway endpoint
@@ -537,7 +537,7 @@ def test_openai_client_with_claude_model_streaming():
     )
 
     stream = client.chat.completions.create(
-        model="claude-sonnet-4-20250514",  # Claude model via OpenAI client
+        model="claude-sonnet-4-6",  # Claude model via OpenAI client
         max_tokens=50,
         messages=[
             {