test: remove end of life model from tests

tests/litellm_utils_tests/test_utils.py

@@ -338,7 +338,9 @@ def test_aget_valid_models():
     print(valid_models)
 
     # list of openai supported llms on litellm
-    expected_models = litellm.open_ai_chat_completion_models | litellm.open_ai_text_completion_models
+    expected_models = (
+        litellm.open_ai_chat_completion_models | litellm.open_ai_text_completion_models
+    )
 
     assert set(valid_models) == set(expected_models)
 
@@ -410,7 +412,12 @@ def test_validate_environment_api_key():
 
 
 def test_validate_environment_api_version():
-    response_obj = validate_environment(model="azure/openai-deployment", api_key="sk-my-test-key", api_base="https://fake.openai.azure.com/", api_version="2024-02-15")
+    response_obj = validate_environment(
+        model="azure/openai-deployment",
+        api_key="sk-my-test-key",
+        api_base="https://fake.openai.azure.com/",
+        api_version="2024-02-15",
+    )
     assert (
         response_obj["keys_in_environment"] is True
     ), f"Missing keys={response_obj['missing_keys']}"
@@ -513,7 +520,6 @@ def test_function_to_dict():
         ("gpt-3.5-turbo", True),
         ("azure/gpt-4-1106-preview", True),
         ("groq/gemma-7b-it", True),
-        ("anthropic.claude-instant-v1", False),
         ("gemini/gemini-1.5-flash", True),
     ],
 )
@@ -1690,15 +1696,6 @@ def test_pick_cheapest_chat_model_from_llm_provider():
     assert len(pick_cheapest_chat_models_from_llm_provider("unknown", n=1)) == 0
 
 
-def test_get_potential_model_names():
-    from litellm.utils import _get_potential_model_names
-
-    assert _get_potential_model_names(
-        model="bedrock/ap-northeast-1/anthropic.claude-instant-v1",
-        custom_llm_provider="bedrock",
-    )
-
-
 @pytest.mark.parametrize("num_retries", [0, 1, 5])
 def test_get_num_retries(num_retries):
     from litellm.utils import _get_wrapper_num_retries

tests/llm_translation/test_bedrock_completion.py

@@ -69,7 +69,7 @@ def test_completion_bedrock_claude_completion_auth():
 
     try:
         response = completion(
-            model="bedrock/anthropic.claude-instant-v1",
+            model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
             messages=messages,
             max_tokens=10,
             temperature=0.1,
@@ -195,7 +195,7 @@ def test_completion_bedrock_claude_external_client_auth():
         )
 
         response = completion(
-            model="bedrock/anthropic.claude-instant-v1",
+            model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
             messages=messages,
             max_tokens=10,
             temperature=0.1,
@@ -232,7 +232,7 @@ def test_completion_bedrock_claude_sts_client_auth():
         litellm.set_verbose = True
 
         response = completion(
-            model="bedrock/anthropic.claude-instant-v1",
+            model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
             messages=messages,
             max_tokens=10,
             temperature=0.1,
@@ -904,7 +904,7 @@ def test_bedrock_ptu():
         )
         try:
             response = litellm.completion(
-                model="bedrock/anthropic.claude-instant-v1",
+                model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
                 messages=[{"role": "user", "content": "What's AWS?"}],
                 model_id=model_id,
                 client=client,
@@ -1070,7 +1070,7 @@ def test_completion_bedrock_external_client_region():
         with patch.object(client, "post", new=Mock()) as mock_client_post:
             try:
                 response = completion(
-                    model="bedrock/anthropic.claude-instant-v1",
+                    model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
                     messages=messages,
                     max_tokens=10,
                     temperature=0.1,

tests/local_testing/test_completion.py

@@ -759,6 +759,7 @@ def test_completion_base64(model):
         else:
             pytest.fail(f"An exception occurred - {str(e)}")
 
+
 def test_completion_mistral_api():
     try:
         litellm.set_verbose = True
@@ -3190,7 +3191,6 @@ def response_format_tests(response: litellm.ModelResponse):
         "bedrock/mistral.mistral-large-2407-v1:0",
         "bedrock/cohere.command-r-plus-v1:0",
         "anthropic.claude-3-sonnet-20240229-v1:0",
-        "anthropic.claude-instant-v1",
         "mistral.mistral-7b-instruct-v0:2",
         # "bedrock/amazon.titan-tg1-large",
         "meta.llama3-8b-instruct-v1:0",

tests/local_testing/test_completion_cost.py

@@ -319,64 +319,9 @@ def test_cost_openai_image_gen():
     assert cost == 0.019922944
 
 
-def test_cost_bedrock_pricing():
-    """
-    - get pricing specific to region for a model
-    """
-    from litellm import Choices, Message, ModelResponse
-    from litellm.utils import Usage
-
-    litellm.set_verbose = True
-    input_tokens = litellm.token_counter(
-        model="bedrock/anthropic.claude-instant-v1",
-        messages=[{"role": "user", "content": "Hey, how's it going?"}],
-    )
-    print(f"input_tokens: {input_tokens}")
-    output_tokens = litellm.token_counter(
-        model="bedrock/anthropic.claude-instant-v1",
-        text="It's all going well",
-        count_response_tokens=True,
-    )
-    print(f"output_tokens: {output_tokens}")
-    resp = ModelResponse(
-        id="chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac",
-        choices=[
-            Choices(
-                finish_reason=None,
-                index=0,
-                message=Message(
-                    content="It's all going well",
-                    role="assistant",
-                ),
-            )
-        ],
-        created=1700775391,
-        model="anthropic.claude-instant-v1",
-        object="chat.completion",
-        system_fingerprint=None,
-        usage=Usage(
-            prompt_tokens=input_tokens,
-            completion_tokens=output_tokens,
-            total_tokens=input_tokens + output_tokens,
-        ),
-    )
-    resp._hidden_params = {
-        "custom_llm_provider": "bedrock",
-        "region_name": "ap-northeast-1",
-    }
-
-    cost = litellm.completion_cost(
-        model="anthropic.claude-instant-v1",
-        completion_response=resp,
-        messages=[{"role": "user", "content": "Hey, how's it going?"}],
-    )
-    predicted_cost = input_tokens * 0.00000223 + 0.00000755 * output_tokens
-    assert cost == predicted_cost
-
-
 def test_cost_bedrock_pricing_actual_calls():
     litellm.set_verbose = True
-    model = "anthropic.claude-instant-v1"
+    model = "anthropic.claude-3-5-sonnet-20240620-v1:0"
     messages = [{"role": "user", "content": "Hey, how's it going?"}]
     response = litellm.completion(
         model=model, messages=messages, mock_response="hello cool one"
@@ -384,7 +329,7 @@ def test_cost_bedrock_pricing_actual_calls():
 
     print("response", response)
     cost = litellm.completion_cost(
-        model="bedrock/anthropic.claude-instant-v1",
+        model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
         completion_response=response,
         messages=[{"role": "user", "content": "Hey, how's it going?"}],
     )
@@ -864,6 +809,7 @@ def test_vertex_ai_embedding_completion_cost(caplog):
 
 #     assert False
 
+
 @pytest.mark.parametrize("sync_mode", [True, False])
 @pytest.mark.asyncio
 async def test_completion_cost_hidden_params(sync_mode):
@@ -949,7 +895,9 @@ def test_vertex_ai_mistral_predict_cost(usage):
     assert predictive_cost > 0
 
 
-@pytest.mark.parametrize("model", ["openai/tts-1", "azure/tts-1", "openai/gpt-4o-mini-tts"])
+@pytest.mark.parametrize(
+    "model", ["openai/tts-1", "azure/tts-1", "openai/gpt-4o-mini-tts"]
+)
 def test_completion_cost_tts(model):
     os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
     litellm.model_cost = litellm.get_model_cost_map(url="")
@@ -1225,7 +1173,10 @@ def test_get_model_params_fireworks_ai(model, base_model):
 
 @pytest.mark.parametrize(
     "model",
-    ["fireworks_ai/llama-v3p1-405b-instruct", "fireworks_ai/llama4-maverick-instruct-basic"],
+    [
+        "fireworks_ai/llama-v3p1-405b-instruct",
+        "fireworks_ai/llama4-maverick-instruct-basic",
+    ],
 )
 def test_completion_cost_fireworks_ai(model):
     os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
@@ -2862,6 +2813,7 @@ def test_cost_calculator_with_custom_pricing():
 @pytest.mark.asyncio
 async def test_cost_calculator_with_custom_pricing_router(model_item, custom_pricing):
     from litellm import Router
+
     if custom_pricing == "litellm_params":
         model_item["litellm_params"]["input_cost_per_token"] = 0.0000008
         model_item["litellm_params"]["output_cost_per_token"] = 0.0000032

tests/local_testing/test_router_timeout.py

@@ -105,7 +105,7 @@ async def test_router_timeouts_bedrock():
         {
             "model_name": "bedrock",
             "litellm_params": {
-                "model": "bedrock/anthropic.claude-instant-v1",
+                "model": "bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
                 "timeout": 0.00001,
             },
             "tpm": 80000,

tests/local_testing/test_streaming.py

@@ -1317,7 +1317,6 @@ async def test_completion_replicate_llama3_streaming(sync_mode):
         # ["bedrock/ai21.jamba-instruct-v1:0", "us-east-1"],
         # ["bedrock/cohere.command-r-plus-v1:0", None],
         ["anthropic.claude-3-sonnet-20240229-v1:0", None],
-        # ["anthropic.claude-instant-v1", None],
         # ["mistral.mistral-7b-instruct-v0:2", None],
         ["bedrock/amazon.titan-tg1-large", None],
         # ["meta.llama3-8b-instruct-v1:0", None],
@@ -1545,50 +1544,6 @@ def test_completion_replicate_stream_bad_key():
 
 # test_completion_replicate_stream_bad_key()
 
-
-def test_completion_bedrock_claude_stream():
-    try:
-        litellm.set_verbose = True
-        response = completion(
-            model="bedrock/anthropic.claude-instant-v1",
-            messages=[
-                {
-                    "role": "user",
-                    "content": "Be as verbose as possible and give as many details as possible, how does a court case get to the Supreme Court?",
-                }
-            ],
-            temperature=1,
-            max_tokens=20,
-            stream=True,
-        )
-        print(response)
-        complete_response = ""
-        has_finish_reason = False
-        # Add any assertions here to check the response
-        first_chunk_id = None
-        for idx, chunk in enumerate(response):
-            # print
-            if idx == 0:
-                first_chunk_id = chunk.id
-            else:
-                assert (
-                    chunk.id == first_chunk_id
-                ), f"chunk ids do not match: {chunk.id} != first chunk id{first_chunk_id}"
-            chunk, finished = streaming_format_tests(idx, chunk)
-            has_finish_reason = finished
-            complete_response += chunk
-            if finished:
-                break
-        if has_finish_reason is False:
-            raise Exception("finish reason not set for last chunk")
-        if complete_response.strip() == "":
-            raise Exception("Empty response received")
-    except RateLimitError:
-        pass
-    except Exception as e:
-        pytest.fail(f"Error occurred: {e}")
-
-
 # test_completion_bedrock_claude_stream()
 
 

tests/local_testing/test_timeout.py

@@ -22,7 +22,6 @@ import litellm
     "model, provider",
     [
         ("gpt-3.5-turbo", "openai"),
-        ("anthropic.claude-instant-v1", "bedrock"),
         ("azure/chatgpt-v-3", "azure"),
     ],
 )
@@ -77,7 +76,7 @@ def test_bedrock_timeout():
     litellm.set_verbose = True
     try:
         response = litellm.completion(
-            model="bedrock/anthropic.claude-instant-v1",
+            model="bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0",
             timeout=0.01,
             messages=[{"role": "user", "content": "hello, write a 20 pg essay"}],
         )