From 4f93e2c3da75289393af1b9e2ddb28935b4e11da Mon Sep 17 00:00:00 2001 From: yuneng-jiang Date: Tue, 22 Sep 2026 17:28:34 -0700 Subject: [PATCH] test: point CircleCI-only suites at models still in the cost map (#42617) * test: point CircleCI-only suites at models still in the cost map #42435 removed cost map entries past their deprecation date and #42437 added litellm_uisettings to the config-synced tables, but both only updated tests/test_litellm. The CircleCI-only suites (local_testing, llm_translation, logging_callback_tests, litellm_utils_tests, unit) kept using the removed models or the old table list and went red on main. Each test keeps its assertions and swaps the removed model for a current one with the same provider and capabilities. The fireworks tests pick a vision model from the cost map because #34941 set supports_vision false on minimax-m3, and the vertex image provider test injects the image model set because #42435 removed every vertex_ai-image-models entry. * test(vertex_ai): register the image model through add_known_models in the provider test --- tests/litellm_utils_tests/test_utils.py | 2 +- .../test_fireworks_ai_translation.py | 14 ++++-- .../test_gemini_image_usage.py | 2 +- tests/llm_translation/test_groq.py | 2 +- tests/llm_translation/test_optional_params.py | 8 +-- tests/llm_translation/test_xai.py | 2 +- .../test_amazing_vertex_completion.py | 6 +-- tests/local_testing/test_completion_cost.py | 50 +++++++++---------- tests/local_testing/test_exceptions.py | 2 +- .../test_function_call_parsing.py | 2 +- tests/local_testing/test_get_llm_provider.py | 18 +++++-- tests/local_testing/test_get_model_info.py | 4 +- .../local_testing/test_lowest_cost_routing.py | 2 +- .../test_openai_moderations_hook.py | 6 +-- tests/local_testing/test_router_utils.py | 24 ++++----- .../test_spend_calculate_endpoint.py | 4 +- .../completion_with_vertex_call.json | 10 ++-- tests/logging_callback_tests/test_alerting.py | 6 +-- .../test_langfuse_e2e_test.py | 4 +- tests/unit/repositories/test_repositories.py | 1 + 20 files changed, 93 insertions(+), 76 deletions(-) diff --git a/tests/litellm_utils_tests/test_utils.py b/tests/litellm_utils_tests/test_utils.py index f7575b969c4..fb20cdf7e0e 100644 --- a/tests/litellm_utils_tests/test_utils.py +++ b/tests/litellm_utils_tests/test_utils.py @@ -263,7 +263,7 @@ def test_trimming_should_not_change_original_messages(): assert messages == messages_copy -@pytest.mark.parametrize("model", ["gpt-4-0125-preview", "claude-sonnet-4-6"]) +@pytest.mark.parametrize("model", ["gpt-5.4-mini", "claude-sonnet-4-6"]) def test_trimming_with_model_cost_max_input_tokens(model): messages = [ {"role": "system", "content": "This is a normal system message"}, diff --git a/tests/llm_translation/test_fireworks_ai_translation.py b/tests/llm_translation/test_fireworks_ai_translation.py index e20134fc1bf..a7dc913c388 100644 --- a/tests/llm_translation/test_fireworks_ai_translation.py +++ b/tests/llm_translation/test_fireworks_ai_translation.py @@ -9,6 +9,12 @@ from litellm.llms.fireworks_ai.chat.transformation import FireworksAIConfig fireworks = FireworksAIConfig() +VISION_MODEL = next( + key.removeprefix("fireworks_ai/") + for key, info in litellm.model_cost.items() + if key.startswith("fireworks_ai/accounts/fireworks/models/") and info.get("supports_vision") is True +) + def test_map_openai_params_tool_choice(): # Test case 1: tool_choice is "required" @@ -97,7 +103,7 @@ def test_document_inlining_example(disable_add_transform_inline_image_block): with patch.object(client, "post") as mock_post: try: completion( - model="fireworks_ai/accounts/fireworks/models/minimax-m3", + model=f"fireworks_ai/{VISION_MODEL}", messages=[ { "role": "user", @@ -157,7 +163,7 @@ def test_transform_inline_no_longer_added(content, expected_url): result = litellm.FireworksAIConfig()._transform_messages_helper( messages=messages, - model="accounts/fireworks/models/minimax-m3", + model=VISION_MODEL, litellm_params={}, ) result_image_block = result[0]["content"][0] @@ -182,7 +188,7 @@ def test_global_disable_flag_no_longer_adds_transform_inline(is_disabled): ] result = litellm.FireworksAIConfig()._transform_messages_helper( messages=messages, - model="accounts/fireworks/models/minimax-m3", + model=VISION_MODEL, litellm_params={}, ) assert result[0]["content"][0]["image_url"] == url @@ -204,7 +210,7 @@ def test_global_disable_flag_with_transform_messages_helper(monkeypatch): ) as mock_post: try: completion( - model="fireworks_ai/accounts/fireworks/models/minimax-m3", + model=f"fireworks_ai/{VISION_MODEL}", messages=[ { "role": "user", diff --git a/tests/llm_translation/test_gemini_image_usage.py b/tests/llm_translation/test_gemini_image_usage.py index 096f9c4796c..0be8b6c23e1 100644 --- a/tests/llm_translation/test_gemini_image_usage.py +++ b/tests/llm_translation/test_gemini_image_usage.py @@ -238,7 +238,7 @@ def test_gemini_image_generation_accumulates_multiple_image_prompt_token_details os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True" litellm.model_cost = litellm.get_model_cost_map(url="") - model = "gemini/gemini-3-pro-image-preview" + model = "gemini/gemini-3-pro-image" config = GoogleImageGenConfig() usage_metadata = { diff --git a/tests/llm_translation/test_groq.py b/tests/llm_translation/test_groq.py index c720f818eaf..fbecbeab08b 100644 --- a/tests/llm_translation/test_groq.py +++ b/tests/llm_translation/test_groq.py @@ -32,7 +32,7 @@ class TestGroq(BaseLLMChatTest): @pytest.mark.parametrize( "model", - ["groq/qwen/qwen3-32b", "groq/openai/gpt-oss-20b", "groq/openai/gpt-oss-120b"], + ["groq/qwen/qwen3.8-27b", "groq/openai/gpt-oss-20b", "groq/openai/gpt-oss-120b"], ) def test_reasoning_effort_in_supported_params(self, model): """Test that reasoning_effort is in the list of supported parameters for Groq""" diff --git a/tests/llm_translation/test_optional_params.py b/tests/llm_translation/test_optional_params.py index 997f5b3b73f..58446014bdf 100644 --- a/tests/llm_translation/test_optional_params.py +++ b/tests/llm_translation/test_optional_params.py @@ -537,7 +537,7 @@ def test_dynamic_drop_params_e2e(): ) as mock_response: try: response = litellm.completion( - model="command-r", + model="command-r-08-2024", messages=[{"role": "user", "content": "Hey, how's it going?"}], response_format={"key": "value"}, drop_params=True, @@ -556,7 +556,7 @@ def test_dynamic_pass_additional_params(): ) as mock_response: try: response = litellm.completion( - model="command-r", + model="command-r-08-2024", messages=[{"role": "user", "content": "Hey, how's it going?"}], custom_param="test", api_key="my-custom-key", @@ -606,7 +606,7 @@ def test_dynamic_drop_params_parallel_tool_calls(): ) as mock_response: try: response = litellm.completion( - model="command-r", + model="command-r-08-2024", messages=[{"role": "user", "content": "Hey, how's it going?"}], parallel_tool_calls=True, drop_params=True, @@ -663,7 +663,7 @@ def test_dynamic_drop_additional_params_e2e(): ) as mock_response: try: response = litellm.completion( - model="command-r", + model="command-r-08-2024", messages=[{"role": "user", "content": "Hey, how's it going?"}], response_format={"key": "value"}, additional_drop_params=["response_format"], diff --git a/tests/llm_translation/test_xai.py b/tests/llm_translation/test_xai.py index 7a121afc3fa..d6d42ed215e 100644 --- a/tests/llm_translation/test_xai.py +++ b/tests/llm_translation/test_xai.py @@ -164,7 +164,7 @@ def test_xai_message_name_filtering(): class TestXAIReasoningEffort(BaseReasoningLLMTests): def get_base_completion_call_args(self): return { - "model": "xai/grok-3-mini-beta", + "model": "xai/grok-4.7", "messages": [{"role": "user", "content": "Hello"}], } diff --git a/tests/local_testing/test_amazing_vertex_completion.py b/tests/local_testing/test_amazing_vertex_completion.py index 3d66064f5c0..8b45ae08813 100644 --- a/tests/local_testing/test_amazing_vertex_completion.py +++ b/tests/local_testing/test_amazing_vertex_completion.py @@ -2863,7 +2863,7 @@ def test_gemini_function_call_parameter_in_messages(): mock_client.return_value = mock_response try: completion( - model="vertex_ai/gemini-2.0-flash", + model="vertex_ai/gemini-2.5-flash-preview-09-2025", messages=messages, tools=tools, tool_choice="auto", @@ -3263,7 +3263,7 @@ def test_vertex_anthropic_completion(): client, "post", side_effect=vertex_ai_anthropic_thinking_mock_response ): response = completion( - model="vertex_ai/claude-3-7-sonnet@20250219", + model="vertex_ai/claude-sonnet-4-6@default", messages=[{"role": "user", "content": "Hello, world!"}], vertex_ai_location="us-east5", vertex_ai_project="test-project", @@ -3271,7 +3271,7 @@ def test_vertex_anthropic_completion(): client=client, ) print(response) - assert response.model == "claude-3-7-sonnet@20250219" + assert response.model == "claude-sonnet-4-6@default" assert response._hidden_params["response_cost"] is not None assert response._hidden_params["response_cost"] > 0 diff --git a/tests/local_testing/test_completion_cost.py b/tests/local_testing/test_completion_cost.py index f40818b9bf1..3ce99f893d8 100644 --- a/tests/local_testing/test_completion_cost.py +++ b/tests/local_testing/test_completion_cost.py @@ -445,7 +445,7 @@ def test_groq_response_cost_tracking(is_streaming): response_cost = litellm.response_cost_calculator( response_object=response, - model="groq/llama-3.3-70b-versatile", + model="groq/openai/gpt-oss-120b", custom_llm_provider="groq", call_type=CallTypes.acompletion.value, optional_params={}, @@ -515,7 +515,7 @@ def test_gemini_completion_cost(provider): """ os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True" litellm.model_cost = litellm.get_model_cost_map(url="") - model_name = "gemini-2.0-flash" + model_name = "gemini-3.8-flash" prompt_tokens = 128.0 output_tokens = 228.0 ## GET MODEL FROM LITELLM.MODEL_INFO @@ -543,7 +543,7 @@ def test_vertex_ai_completion_cost(): prompt_tokens = 100 - model_info = litellm.get_model_info(model="gemini-2.0-flash") + model_info = litellm.get_model_info(model="gemini-3.8-flash") print("\nExpected model info:\n{}\n\n".format(model_info)) @@ -551,7 +551,7 @@ def test_vertex_ai_completion_cost(): ## CALCULATED COST calculated_input_cost, calculated_output_cost = cost_per_token( - model="gemini-2.0-flash", + model="gemini-3.8-flash", custom_llm_provider="vertex_ai", prompt_tokens=prompt_tokens, completion_tokens=0, @@ -676,7 +676,7 @@ async def test_completion_cost_hidden_params(sync_mode): def test_vertex_ai_gemini_predict_cost(): - model = "gemini-2.0-flash" + model = "gemini-3.8-flash" messages = [{"role": "user", "content": "Hey, hows it going???"}] predictive_cost = completion_cost(model=model, messages=messages) @@ -757,24 +757,24 @@ def test_completion_cost_tts(model): def test_completion_cost_anthropic(): """ - model_name: claude-3-haiku-20240307 + model_name: claude-haiku-4-5 litellm_params: - model: anthropic/claude-3-haiku-20240307 + model: anthropic/claude-haiku-4-5 max_tokens: 4096 """ router = litellm.Router( model_list=[ { - "model_name": "claude-3-haiku-20240307", + "model_name": "claude-haiku-4-5", "litellm_params": { - "model": "anthropic/claude-3-haiku-20240307", + "model": "anthropic/claude-haiku-4-5", "max_tokens": 4096, }, } ] ) data = { - "model": "claude-3-haiku-20240307", + "model": "claude-haiku-4-5", "prompt_tokens": 21, "completion_tokens": 20, "response_time_ms": 871.7040000000001, @@ -2068,14 +2068,14 @@ def test_completion_cost_params(): """ litellm.set_verbose = True resp1_prompt_cost, resp1_completion_cost = cost_per_token( - model="gemini-2.0-flash", + model="gemini-3.8-flash", prompt_tokens=1000, completion_tokens=1000, custom_llm_provider="vertex_ai_beta", ) resp2_prompt_cost, resp2_completion_cost = cost_per_token( - model="gemini-2.0-flash", prompt_tokens=1000, completion_tokens=1000 + model="gemini-3.8-flash", prompt_tokens=1000, completion_tokens=1000 ) assert resp2_prompt_cost > 0 @@ -2084,7 +2084,7 @@ def test_completion_cost_params(): assert resp1_completion_cost == resp2_completion_cost resp3_prompt_cost, resp3_completion_cost = cost_per_token( - model="vertex_ai/gemini-2.0-flash", prompt_tokens=1000, completion_tokens=1000 + model="vertex_ai/gemini-3.8-flash", prompt_tokens=1000, completion_tokens=1000 ) assert resp3_prompt_cost > 0 @@ -2102,14 +2102,14 @@ def test_completion_cost_params_2(): prompt_tokens = 1000 completion_tokens = 1000 resp1_prompt_cost, resp1_completion_cost = cost_per_token( - model="gemini-2.0-flash", + model="gemini-3.8-flash", prompt_tokens=prompt_tokens, completion_tokens=completion_tokens, ) print(resp1_prompt_cost, resp1_completion_cost) - model_info = litellm.get_model_info("gemini-2.0-flash") + model_info = litellm.get_model_info("gemini-3.8-flash") input_cost_per_token = model_info["input_cost_per_token"] output_cost_per_token = model_info["output_cost_per_token"] @@ -2148,7 +2148,7 @@ def test_completion_cost_params_gemini_3(): ) ], created=1728529259, - model="gemini-2.0-flash", + model="gemini-3.8-flash", object="chat.completion", system_fingerprint=None, usage=usage, @@ -2172,7 +2172,7 @@ def test_completion_cost_params_gemini_3(): pc, cc = cost_per_character( **{ - "model": "gemini-2.0-flash", + "model": "gemini-3.8-flash", "custom_llm_provider": "vertex_ai", "prompt_characters": None, "completion_characters": 3, @@ -2180,9 +2180,9 @@ def test_completion_cost_params_gemini_3(): } ) - model_info = litellm.get_model_info("gemini-2.0-flash") + model_info = litellm.get_model_info("gemini-3.8-flash") - # gemini-2.0-flash has no per-character pricing, so cost_per_character + # gemini-3.8-flash has no per-character pricing, so cost_per_character # falls back to per-token pricing using usage.prompt_tokens / usage.completion_tokens assert round(pc, 10) == round(3771 * model_info["input_cost_per_token"], 10) assert round(cc, 10) == round( @@ -2239,16 +2239,16 @@ async def test_test_completion_cost_gpt4o_audio_output_from_model(stream): ) ], created=1729282652, - model="gpt-4o-audio-preview", + model="gpt-audio-1.5", object="chat.completion", system_fingerprint="fp_4eafc16e9d", usage=usage_object, service_tier=None, ) - cost = completion_cost(completion, model="gpt-4o-audio-preview") + cost = completion_cost(completion, model="gpt-audio-1.5") - model_info = litellm.get_model_info("gpt-4o-audio-preview") + model_info = litellm.get_model_info("gpt-audio-1.5") print(f"model_info: {model_info}") ## input cost @@ -2517,7 +2517,7 @@ def test_cost_calculator_with_base_model(): resp = litellm.completion( model="bedrock/random-model", messages=[{"role": "user", "content": "Hello, how are you?"}], - base_model="bedrock/anthropic.claude-3-sonnet-20240229-v1:0", + base_model="bedrock/anthropic.claude-sonnet-5", mock_response="Hello, how are you?", ) assert resp.model == "random-model" @@ -2551,10 +2551,10 @@ def test_cost_calculator_with_base_model_with_router(base_model_arg): if base_model_arg == "litellm_param": model_item["litellm_params"][ "base_model" - ] = "bedrock/anthropic.claude-3-sonnet-20240229-v1:0" + ] = "bedrock/anthropic.claude-sonnet-5" elif base_model_arg == "model_info": model_item["model_info"] = { - "base_model": "bedrock/anthropic.claude-3-sonnet-20240229-v1:0", + "base_model": "bedrock/anthropic.claude-sonnet-5", } router = Router(model_list=[model_item]) diff --git a/tests/local_testing/test_exceptions.py b/tests/local_testing/test_exceptions.py index e6392cda406..813146f8ace 100644 --- a/tests/local_testing/test_exceptions.py +++ b/tests/local_testing/test_exceptions.py @@ -1148,7 +1148,7 @@ def test_openai_gateway_timeout_error(): @pytest.mark.parametrize( "provider, model, call_type", [ - ("anthropic", "claude-3-haiku-20240307", "chat_completion"), + ("anthropic", "claude-haiku-4-5-20251001", "chat_completion"), ], ) @pytest.mark.asyncio diff --git a/tests/local_testing/test_function_call_parsing.py b/tests/local_testing/test_function_call_parsing.py index c98f170a98f..ebb13e0018d 100644 --- a/tests/local_testing/test_function_call_parsing.py +++ b/tests/local_testing/test_function_call_parsing.py @@ -136,7 +136,7 @@ def trade(model_name: str) -> List[Trade]: # type: ignore @pytest.mark.parametrize( - "model", ["claude-haiku-4-5-20251001", "anthropic.claude-3-haiku-20240307-v1:0"] + "model", ["claude-haiku-4-5-20251001", "us.anthropic.claude-haiku-4-5-20251001-v1:0"] ) @pytest.mark.flaky(retries=6, delay=10) def test_function_call_parsing(model): diff --git a/tests/local_testing/test_get_llm_provider.py b/tests/local_testing/test_get_llm_provider.py index ebad0fbafc5..4ac7cecb97a 100644 --- a/tests/local_testing/test_get_llm_provider.py +++ b/tests/local_testing/test_get_llm_provider.py @@ -67,7 +67,17 @@ def test_get_llm_provider_deepseek_custom_api_base(): os.environ.pop("DEEPSEEK_API_BASE") -def test_get_llm_provider_vertex_ai_image_models(): +def test_get_llm_provider_vertex_ai_image_models(monkeypatch): + monkeypatch.setattr(litellm, "vertex_ai_image_models", set()) + monkeypatch.setattr(litellm, "models_by_provider", dict(litellm.models_by_provider)) + litellm.add_known_models( + model_cost_map={ + "vertex_ai/imagegeneration@006": { + "litellm_provider": "vertex_ai-image-models", + "mode": "image_generation", + } + } + ) model, custom_llm_provider, dynamic_api_key, api_base = litellm.get_llm_provider( model="imagegeneration@006", custom_llm_provider=None ) @@ -101,17 +111,17 @@ def test_get_llm_provider_ai21_chat_test2(): def test_get_llm_provider_cohere_chat_test2(): """ - if user prefix with cohere/ but calls command-r-plus then it should be cohere_chat provider + if user prefix with cohere/ but calls command-r-plus-08-2024 then it should be cohere_chat provider """ model, custom_llm_provider, dynamic_api_key, api_base = litellm.get_llm_provider( - model="cohere/command-r-plus", + model="cohere/command-r-plus-08-2024", ) print("model=", model) print("custom_llm_provider=", custom_llm_provider) print("api_base=", api_base) assert custom_llm_provider == "cohere_chat" - assert model == "command-r-plus" + assert model == "command-r-plus-08-2024" def test_get_llm_provider_azure_o1(): diff --git a/tests/local_testing/test_get_model_info.py b/tests/local_testing/test_get_model_info.py index 37f4ece611d..1e46a1bf853 100644 --- a/tests/local_testing/test_get_model_info.py +++ b/tests/local_testing/test_get_model_info.py @@ -16,7 +16,7 @@ def test_get_model_info_simple_model_name(): """ tests if model name given, and model exists in model info - the object is returned """ - model = "claude-3-opus-20240229" + model = "claude-opus-5-5" litellm.get_model_info(model) @@ -24,7 +24,7 @@ def test_get_model_info_custom_llm_with_model_name(): """ Tests if {custom_llm_provider}/{model_name} name given, and model exists in model info, the object is returned """ - model = "anthropic/claude-3-opus-20240229" + model = "anthropic/claude-opus-5-5" litellm.get_model_info(model) diff --git a/tests/local_testing/test_lowest_cost_routing.py b/tests/local_testing/test_lowest_cost_routing.py index 5bf3a3ee98b..631271ca710 100644 --- a/tests/local_testing/test_lowest_cost_routing.py +++ b/tests/local_testing/test_lowest_cost_routing.py @@ -28,7 +28,7 @@ async def test_get_available_deployments(): }, { "model_name": "gpt-3.5-turbo", - "litellm_params": {"model": "groq/llama-3.1-8b-instant"}, + "litellm_params": {"model": "groq/openai/gpt-oss-20b"}, "model_info": {"id": "groq-llama"}, }, ] diff --git a/tests/local_testing/test_openai_moderations_hook.py b/tests/local_testing/test_openai_moderations_hook.py index 530ab714eae..7ce4bc2e4bf 100644 --- a/tests/local_testing/test_openai_moderations_hook.py +++ b/tests/local_testing/test_openai_moderations_hook.py @@ -31,7 +31,7 @@ async def test_openai_moderation_error_raising(monkeypatch): from unittest.mock import AsyncMock, MagicMock from litellm.types.llms.openai import OpenAIModerationResponse - litellm.openai_moderations_model_name = "text-moderation-latest" + litellm.openai_moderations_model_name = "omni-moderation-latest" openai_mod = _ENTERPRISE_OpenAI_Moderation() _api_key = "sk-12345" _api_key = hash_token("sk-12345") @@ -41,9 +41,9 @@ async def test_openai_moderation_error_raising(monkeypatch): llm_router = litellm.Router( model_list=[ { - "model_name": "text-moderation-latest", + "model_name": "omni-moderation-latest", "litellm_params": { - "model": "text-moderation-latest", + "model": "omni-moderation-latest", "api_key": os.environ.get("OPENAI_API_KEY", "fake-key"), }, } diff --git a/tests/local_testing/test_router_utils.py b/tests/local_testing/test_router_utils.py index 1b3e361bb1f..635bda55144 100644 --- a/tests/local_testing/test_router_utils.py +++ b/tests/local_testing/test_router_utils.py @@ -188,7 +188,7 @@ def test_router_get_model_info_wildcard_routes(): ] ) model_info = router.get_router_model_info( - deployment=None, received_model_name="gemini/gemini-1.5-flash", id="1" + deployment=None, received_model_name="gemini/gemini-2.5-flash", id="1" ) print(model_info) assert model_info is not None @@ -212,7 +212,7 @@ async def test_router_get_model_group_usage_wildcard_routes(): ) resp = await router.acompletion( - model="gemini/gemini-1.5-flash", + model="gemini/gemini-2.5-flash", messages=[{"role": "user", "content": "Hello, how are you?"}], mock_response="Hello, I'm good.", ) @@ -220,7 +220,7 @@ async def test_router_get_model_group_usage_wildcard_routes(): await asyncio.sleep(2) - tpm, rpm = await router.get_model_group_usage(model_group="gemini/gemini-1.5-flash") + tpm, rpm = await router.get_model_group_usage(model_group="gemini/gemini-2.5-flash") assert tpm is not None, "tpm is None" assert rpm is not None, "rpm is None" @@ -242,7 +242,7 @@ async def test_call_router_callbacks_on_success(): router.cache, "async_increment_cache_pipeline", new=AsyncMock() ) as mock_callback: await router.acompletion( - model="gemini/gemini-1.5-flash", + model="gemini/gemini-2.5-flash", messages=[{"role": "user", "content": "Hello, how are you?"}], mock_response="Hello, I'm good.", ) @@ -255,12 +255,12 @@ async def test_call_router_callbacks_on_success(): for increment in increment_list: if "tpm" in increment["key"]: assert increment["key"].startswith( - "global_router:1:gemini/gemini-1.5-flash:tpm" + "global_router:1:gemini/gemini-2.5-flash:tpm" ) assert increment["increment_value"] == 30 elif "rpm" in increment["key"]: assert increment["key"].startswith( - "global_router:1:gemini/gemini-1.5-flash:rpm" + "global_router:1:gemini/gemini-2.5-flash:rpm" ) assert increment["increment_value"] == 1 @@ -283,7 +283,7 @@ async def test_call_router_callbacks_on_failure(): ) as mock_callback: with pytest.raises(litellm.RateLimitError): await router.acompletion( - model="gemini/gemini-1.5-flash", + model="gemini/gemini-2.5-flash", messages=[{"role": "user", "content": "Hello, how are you?"}], mock_response="litellm.RateLimitError", num_retries=0, @@ -295,7 +295,7 @@ async def test_call_router_callbacks_on_failure(): assert ( mock_callback.call_args_list[0] .kwargs["key"] - .startswith("global_router:1:gemini/gemini-1.5-flash:rpm") + .startswith("global_router:1:gemini/gemini-2.5-flash:rpm") ) @@ -317,7 +317,7 @@ async def test_router_model_group_headers(): for _ in range(2): resp = await router.acompletion( - model="gemini/gemini-1.5-flash", + model="gemini/gemini-2.5-flash", messages=[{"role": "user", "content": "Hello, how are you?"}], mock_response="Hello, I'm good.", ) @@ -325,7 +325,7 @@ async def test_router_model_group_headers(): assert ( resp._hidden_params["additional_headers"]["x-litellm-model-group"] - == "gemini/gemini-1.5-flash" + == "gemini/gemini-2.5-flash" ) assert "x-ratelimit-remaining-requests" in resp._hidden_params["additional_headers"] @@ -349,7 +349,7 @@ async def test_get_remaining_model_group_usage(): ) for _ in range(2): resp = await router.acompletion( - model="gemini/gemini-1.5-flash", + model="gemini/gemini-2.5-flash", messages=[{"role": "user", "content": "Hello, how are you?"}], mock_response="Hello, I'm good.", ) @@ -363,7 +363,7 @@ async def test_get_remaining_model_group_usage(): await asyncio.sleep(1) remaining_usage = await router.get_remaining_model_group_usage( - model_group="gemini/gemini-1.5-flash" + model_group="gemini/gemini-2.5-flash" ) assert remaining_usage is not None assert "x-ratelimit-remaining-requests" in remaining_usage diff --git a/tests/local_testing/test_spend_calculate_endpoint.py b/tests/local_testing/test_spend_calculate_endpoint.py index 3bedab794e2..054dc398039 100644 --- a/tests/local_testing/test_spend_calculate_endpoint.py +++ b/tests/local_testing/test_spend_calculate_endpoint.py @@ -38,7 +38,7 @@ async def test_spend_calc_model_on_router_messages(): { "model_name": "special-llama-model", "litellm_params": { - "model": "groq/llama-3.1-8b-instant", + "model": "groq/openai/gpt-oss-20b", }, } ] @@ -81,7 +81,7 @@ async def test_spend_calc_using_response(): } ], "created": "1677652288", - "model": "groq/llama-3.1-8b-instant", + "model": "groq/openai/gpt-oss-20b", "object": "chat.completion", "system_fingerprint": "fp_873a560973", "usage": { diff --git a/tests/logging_callback_tests/langfuse_expected_request_body/completion_with_vertex_call.json b/tests/logging_callback_tests/langfuse_expected_request_body/completion_with_vertex_call.json index b6c11f96953..5998c52659c 100644 --- a/tests/logging_callback_tests/langfuse_expected_request_body/completion_with_vertex_call.json +++ b/tests/logging_callback_tests/langfuse_expected_request_body/completion_with_vertex_call.json @@ -31,14 +31,14 @@ "model_id": null, "cache_key": null, "api_base": null, - "response_cost": 7.5e-06, + "response_cost": 3.5e-05, "additional_headers": {}, "litellm_overhead_time_ms": null, "batch_models": null, - "litellm_model_name": "vertex_ai/gemini-2.0-flash-001", + "litellm_model_name": "vertex_ai/gemini-3-flash-preview", "usage_object": null }, - "litellm_response_cost": 7.5e-06, + "litellm_response_cost": 3.5e-05, "cache_hit": false, "requester_metadata": {} }, @@ -54,13 +54,13 @@ "id": "time-14-15-40-349639_chatcmpl-59a988d0-7ef1-4dc4-bc18-d2e78961817f", "endTime": "2025-05-26T14:15:40.607266-07:00", "completionStartTime": "2025-05-26T14:15:40.607266-07:00", - "model": "gemini-2.0-flash-001", + "model": "gemini-3-flash-preview", "modelParameters": {}, "usage": { "input": 10, "output": 10, "unit": "TOKENS", - "totalCost": 7.5e-06 + "totalCost": 3.5e-05 }, "usageDetails": { "input": 10, diff --git a/tests/logging_callback_tests/test_alerting.py b/tests/logging_callback_tests/test_alerting.py index 3074e973a8e..0a3e1a0e982 100644 --- a/tests/logging_callback_tests/test_alerting.py +++ b/tests/logging_callback_tests/test_alerting.py @@ -582,7 +582,7 @@ async def test_webhook_alerting(alerting_type): None, None, ), - ("gemini-2.0-flash", None, "vertex_ai", "hardy-device-38811", "us-central1"), + ("gemini-3.8-flash", None, "vertex_ai", "hardy-device-38811", "us-central1"), ], ) @pytest.mark.parametrize("error_code", [500, 408, 400]) @@ -688,7 +688,7 @@ async def test_outage_alerting_called( None, None, ), - ("gemini-2.0-flash", None, "vertex_ai", "hardy-device-38811", "us-central1"), + ("gemini-3.8-flash", None, "vertex_ai", "hardy-device-38811", "us-central1"), ], ) @pytest.mark.parametrize("error_code", [500, 408, 400]) @@ -775,7 +775,7 @@ async def test_region_outage_alerting_called( await slack_alerting.region_outage_alerts( exception=error_to_raise, deployment_id=deployment_id # type: ignore ) - if model == "gemini-2.0-flash" and (error_code == 500 or error_code == 408): + if model == "gemini-3.8-flash" and (error_code == 500 or error_code == 408): mock_send_alert.assert_called_once() else: mock_send_alert.assert_not_called() diff --git a/tests/logging_callback_tests/test_langfuse_e2e_test.py b/tests/logging_callback_tests/test_langfuse_e2e_test.py index 5682d3720d8..76ebd2b9a28 100644 --- a/tests/logging_callback_tests/test_langfuse_e2e_test.py +++ b/tests/logging_callback_tests/test_langfuse_e2e_test.py @@ -481,12 +481,12 @@ class TestLangfuseLogging: completion_tokens=10, total_tokens=20, ), - model="vertex/gemini-2.0-flash-001", + model="vertex/gemini-3-flash-preview", object="chat.completion", created=1723081200, ).model_dump() await litellm.acompletion( - model="vertex_ai/gemini-2.0-flash-001", + model="vertex_ai/gemini-3-flash-preview", messages=[{"role": "user", "content": "Hello!"}], mock_response=mock_response, metadata={"trace_id": setup["trace_id"]}, diff --git a/tests/unit/repositories/test_repositories.py b/tests/unit/repositories/test_repositories.py index 87cf2fc4268..e185d95ffb8 100644 --- a/tests/unit/repositories/test_repositories.py +++ b/tests/unit/repositories/test_repositories.py @@ -2120,6 +2120,7 @@ class TestPrismaTableRepository: "litellm_prompttable", "litellm_searchtoolstable", "litellm_ssoconfig", + "litellm_uisettings", } )