[ROCm][CI] Fix flaky Cohere/OpenAI embedding parity test (#37616)

Signed-off-by: Andreas Karatzas <akaratza@amd.com>

tests/entrypoints/pooling/embed/test_cohere_openai_parity.py

@@ -10,7 +10,7 @@ import numpy as np
 import pytest
 import requests
 
-from tests.utils import RemoteOpenAIServer
+from tests.utils import ROCM_EXTRA_ARGS, RemoteOpenAIServer
 
 MODEL_NAME = "BAAI/bge-base-en-v1.5"
 DTYPE = "bfloat16"
@@ -28,7 +28,7 @@ def server():
         "512",
         "--gpu-memory-utilization",
         "0.02",
-    ]
+    ] + ROCM_EXTRA_ARGS
     with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
         yield remote_server
 

tests/entrypoints/pooling/embed/test_online_dimensions.py

@@ -10,7 +10,7 @@ import pytest
 from tests.conftest import HfRunner
 from tests.models.language.pooling.embed_utils import run_embedding_correctness_test
 from tests.models.utils import EmbedModelInfo
-from tests.utils import RemoteOpenAIServer
+from tests.utils import ROCM_EXTRA_ARGS, RemoteOpenAIServer
 from vllm.entrypoints.pooling.embed.protocol import EmbeddingResponse
 from vllm.platforms import current_platform
 
@@ -49,7 +49,7 @@ def server(model_info, dtype: str):
         "--enforce-eager",
         "--max-model-len",
         "512",
-    ]
+    ] + ROCM_EXTRA_ARGS
 
     if model_info.name == "Snowflake/snowflake-arctic-embed-m-v1.5":
         # Manually enable Matryoshka Embeddings

vllm/entrypoints/pooling/base/serving.py

@@ -118,6 +118,7 @@ class PoolingServing:
         )
 
         pooling_params = self.io_processor.create_pooling_params(ctx.request)
+        pooling_params.verify(self.model_config)
 
         for i, engine_prompt in enumerate(ctx.engine_prompts):
             prompt_request_id = (

vllm/entrypoints/utils.py

@@ -309,6 +309,9 @@ def create_error_response(
 
     if isinstance(message, Exception):
         exc = message
+        logger.debug(
+            "create_error_response called with %s: %s", type(exc).__name__, exc
+        )
 
         from vllm.exceptions import VLLMNotFoundError, VLLMValidationError