From 01e1919489c49be9244294bfeec6581da7cb24aa Mon Sep 17 00:00:00 2001 From: mateo-berri <277851410+mateo-berri@users.noreply.github.com> Date: Fri, 30 Oct 2026 09:24:54 -0700 Subject: [PATCH] test(e2e): move audio and realtime tests off OpenAI model ids with shutdown dates --- .../e2e/llm_translation/realtime/REALTIME_COVERAGE_MATRIX.md | 2 +- tests/e2e/llm_translation/realtime/realtime_client.py | 2 +- tests/e2e/llm_translation/realtime/test_realtime_e2e.py | 2 +- tests/e2e/llm_translation/test_audio_speech_e2e.py | 2 +- tests/e2e/llm_translation/test_audio_transcriptions_e2e.py | 2 +- tests/e2e/llm_translation/test_realtime_http_e2e.py | 2 +- .../logging/test_otel_v2_langfuse_generation_output_e2e.py | 4 ++-- 7 files changed, 8 insertions(+), 8 deletions(-) diff --git a/tests/e2e/llm_translation/realtime/REALTIME_COVERAGE_MATRIX.md b/tests/e2e/llm_translation/realtime/REALTIME_COVERAGE_MATRIX.md index c85471da90d..5f26afe86d2 100644 --- a/tests/e2e/llm_translation/realtime/REALTIME_COVERAGE_MATRIX.md +++ b/tests/e2e/llm_translation/realtime/REALTIME_COVERAGE_MATRIX.md @@ -40,7 +40,7 @@ at call time. The provider table below is the source of truth; edit `PROVIDERS` | provider | model alias | upstream model | |----------|-------------|----------------| | openai | `openai-realtime` | `openai/gpt-realtime-2` | -| azure | `azure-realtime` | `azure/gpt-realtime-2` (GA protocol) | +| azure | `azure-realtime` | `azure/gpt-realtime-1.5` (GA protocol) | | gemini | `gemini-realtime` | `gemini/gemini-3.1-flash-live-preview` | | vertex_ai | `vertex-realtime` | `vertex_ai/gemini-live-2.5-flash-native-audio` | diff --git a/tests/e2e/llm_translation/realtime/realtime_client.py b/tests/e2e/llm_translation/realtime/realtime_client.py index 6bd5d61d721..3541fb4386d 100644 --- a/tests/e2e/llm_translation/realtime/realtime_client.py +++ b/tests/e2e/llm_translation/realtime/realtime_client.py @@ -63,7 +63,7 @@ PROVIDERS = ( "azure", "azure-realtime", LiteLLMParamsBody( - model="azure/gpt-realtime", + model="azure/gpt-realtime-1.5", api_key="os.environ/AZURE_API_KEY", api_version="2025-08-28", realtime_protocol="GA", diff --git a/tests/e2e/llm_translation/realtime/test_realtime_e2e.py b/tests/e2e/llm_translation/realtime/test_realtime_e2e.py index 3b7ba3a2430..f0cc358c028 100644 --- a/tests/e2e/llm_translation/realtime/test_realtime_e2e.py +++ b/tests/e2e/llm_translation/realtime/test_realtime_e2e.py @@ -46,7 +46,7 @@ from websockets.exceptions import ConnectionClosedError pytestmark = pytest.mark.e2e -AZURE_REALTIME_MODEL: Final = "azure/gpt-realtime" +AZURE_REALTIME_MODEL: Final = "azure/gpt-realtime-1.5" TEXT_PARAMS: Final = tuple( pytest.param( diff --git a/tests/e2e/llm_translation/test_audio_speech_e2e.py b/tests/e2e/llm_translation/test_audio_speech_e2e.py index c4c3f751828..8f96e0fdd1f 100644 --- a/tests/e2e/llm_translation/test_audio_speech_e2e.py +++ b/tests/e2e/llm_translation/test_audio_speech_e2e.py @@ -24,7 +24,7 @@ from sdk_clients import SdkClients, response_header pytestmark = pytest.mark.e2e -OPENAI_TTS_MODEL: Final = "openai/gpt-4o-mini-tts" +OPENAI_TTS_MODEL: Final = "openai/tts-1-1106" AWS_POLLY_MODEL: Final = "aws_polly/generative" diff --git a/tests/e2e/llm_translation/test_audio_transcriptions_e2e.py b/tests/e2e/llm_translation/test_audio_transcriptions_e2e.py index cd1ee8fb03c..654be4132b7 100644 --- a/tests/e2e/llm_translation/test_audio_transcriptions_e2e.py +++ b/tests/e2e/llm_translation/test_audio_transcriptions_e2e.py @@ -31,7 +31,7 @@ WEATHER_WAV = ( Path(__file__).resolve().parent / "realtime" / "fixtures" / "weather_question_24k.wav" ) -OPENAI_TRANSCRIBE_MODEL: Final = "openai/gpt-4o-mini-transcribe" +OPENAI_TRANSCRIBE_MODEL: Final = "openai/gpt-transcribe" OPENAI_WHISPER_MODEL: Final = "openai/whisper-1" MISSING_MODEL_PHRASES: Final = ("model=none", "invalid model", "model is required") diff --git a/tests/e2e/llm_translation/test_realtime_http_e2e.py b/tests/e2e/llm_translation/test_realtime_http_e2e.py index ee3ce3893c2..bc05fd658c0 100644 --- a/tests/e2e/llm_translation/test_realtime_http_e2e.py +++ b/tests/e2e/llm_translation/test_realtime_http_e2e.py @@ -17,7 +17,7 @@ from pydantic import BaseModel pytestmark = pytest.mark.e2e -REALTIME_BACKEND = "openai/gpt-realtime" +REALTIME_BACKEND = "openai/gpt-realtime-2.1" class RealtimeSession(BaseModel): diff --git a/tests/e2e/logging/test_otel_v2_langfuse_generation_output_e2e.py b/tests/e2e/logging/test_otel_v2_langfuse_generation_output_e2e.py index 8aa5f680972..c0b648c9810 100644 --- a/tests/e2e/logging/test_otel_v2_langfuse_generation_output_e2e.py +++ b/tests/e2e/logging/test_otel_v2_langfuse_generation_output_e2e.py @@ -72,8 +72,8 @@ BOUNDED_OUTPUT_CHARS: Final = 1024 PLACEHOLDER_INPUT: Final = "default-message-value" COMPLETION_BACKEND: Final = "openai/gpt-3.5-turbo-instruct" IMAGE_BACKEND: Final = "openai/gpt-image-1-mini" -SPEECH_BACKEND: Final = "openai/gpt-4o-mini-tts" -TRANSCRIPTION_BACKEND: Final = "openai/gpt-4o-mini-transcribe" +SPEECH_BACKEND: Final = "openai/tts-1-1106" +TRANSCRIPTION_BACKEND: Final = "openai/gpt-transcribe" MODERATION_BACKEND: Final = "openai/omni-moderation-latest" MISTRAL_OCR_BACKEND: Final = "mistral/mistral-ocr-latest" RERANK_BACKEND: Final = "cohere/rerank-v4.0-fast"