[Bug]: model_executor/test_model_load_with_params.py fails with AttributeError #18757
Description
Your current environment
Issue encountered on main branch tests.
🐛 Describe the bug
Test failing with below traceback:
vllm_runner = <class 'tests.conftest.VllmRunner'>
@pytest.mark.skipif(current_platform.is_rocm(),
reason="Xformers backend is not supported on ROCm.")
def test_model_loading_with_params(vllm_runner):
"""
Test parameter weight loading with tp>1.
"""
with vllm_runner(model_name=MODEL_NAME,
revision=REVISION,
dtype="float16",
max_model_len=MAX_MODEL_LEN) as vllm_model:
output = vllm_model.encode("Write a short story about a robot that"
" dreams for the first time.\n")
model_config = vllm_model.model.llm_engine.model_config
model_tokenizer = vllm_model.model.llm_engine.tokenizer
# asserts on the bert model config file
assert model_config.encoder_config["max_seq_length"] == 512
assert model_config.encoder_config["do_lower_case"]
# asserts on the pooling config files
assert model_config.pooler_config.pooling_type == PoolingType.CLS.name
> assert model_config.pooler_config.pooling_norm
E AttributeError: 'PoolerConfig' object has no attribute 'pooling_norm'. Did you mean: 'pooling_type'?
tests/model_executor/test_model_load_with_params.py:43: AttributeError
------------------------------------------------- Captured stdout call -------------------------------------------------
INFO 05-27 12:31:33 [__init__.py:31] Available plugins for group vllm.general_plugins:
INFO 05-27 12:31:33 [__init__.py:33] - lora_filesystem_resolver -> vllm.plugins.lora_resolvers.filesystem_resolver:register_filesystem_resolver
INFO 05-27 12:31:33 [__init__.py:36] All plugins in this group will be loaded. Set `VLLM_PLUGINS` to control which plugins to load.
INFO 05-27 12:32:10 [config.py:577] Found sentence-transformers tokenize configuration.
INFO 05-27 12:32:10 [config.py:3131] Downcasting torch.float32 to torch.float16.
INFO 05-27 12:32:18 [config.py:473] Found sentence-transformers modules configuration.
INFO 05-27 12:32:18 [config.py:493] Found pooling configuration.
INFO 05-27 12:32:18 [config.py:793] This model supports multiple tasks: {'reward', 'embed', 'classify', 'score'}. Defaulting to 'embed'.
WARNING 05-27 12:32:18 [arg_utils.py:1583] --task embed is not supported by the V1 Engine. Falling back to V0.
INFO 05-27 12:32:18 [llm_engine.py:230] Initializing a V0 LLM engine (v0.9.1.dev148+gfba064270.d20250526) with config: model='BAAI/bge-base-en-v1.5', speculative_config=None, tokenizer='BAAI/bge-base-en-v1.5', skip_tokenizer_init=False, tokenizer_mode=auto, revision=main, override_neuron_config={}, tokenizer_revision=main, trust_remote_code=True, dtype=torch.float16, max_seq_len=128, download_dir=None, load_format=LoadFormat.AUTO, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto, device_config=cuda, decoding_config=DecodingConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_backend=''), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None), seed=0, served_model_name=BAAI/bge-base-en-v1.5, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=None, chunked_prefill_enabled=False, use_async_output_proc=False, pooler_config=PoolerConfig(pooling_type='CLS', normalize=True, softmax=None, step_tag_id=None, returned_token_ids=None), compilation_config={"compile_sizes": [], "inductor_compile_config": {"enable_auto_functionalized_v2": false}, "cudagraph_capture_sizes": [256, 248, 240, 232, 224, 216, 208, 200, 192, 184, 176, 168, 160, 152, 144, 136, 128, 120, 112, 104, 96, 88, 80, 72, 64, 56, 48, 40, 32, 24, 16, 8, 4, 2, 1], "max_capture_size": 256}, use_cached_outputs=False,
INFO 05-27 12:32:18 [cuda.py:292] Using Flash Attention backend.
INFO 05-27 12:32:19 [parallel_state.py:1064] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, TP rank 0, EP rank 0
INFO 05-27 12:32:19 [model_runner.py:1170] Starting to load model BAAI/bge-base-en-v1.5...
INFO 05-27 12:32:19 [weight_utils.py:291] Using model weights format ['*.safetensors']
INFO 05-27 12:32:20 [weight_utils.py:344] No model.safetensors.index.json found in remote.
INFO 05-27 12:32:20 [default_loader.py:280] Loading weights took 0.08 seconds
INFO 05-27 12:32:20 [model_runner.py:1202] Model loading took 0.2091 GiB and 0.738950 seconds
------------------------------------------------- Captured stderr call -------------------------------------------------
Loading safetensors checkpoint shards: 0% Completed | 0/1 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:00<00:00, 13.57it/s]
Adding requests: 100%|██████████| 66/66 [00:00<00:00, 3031.89it/s]
Processed prompts: 100%|██████████| 66/66 [00:00<00:00, 176.21it/s, est. speed input: 0.00 toks/s, output: 0.00 toks/s]
________________________________________ test_roberta_model_loading_with_params ________________________________________
vllm_runner = <class 'tests.conftest.VllmRunner'>
@pytest.mark.skipif(current_platform.is_rocm(),
reason="Xformers backend is not supported on ROCm.")
def test_roberta_model_loading_with_params(vllm_runner):
"""
Test parameter weight loading with tp>1.
"""
with vllm_runner(model_name=MODEL_NAME_ROBERTA,
revision=REVISION_ROBERTA,
dtype="float16",
max_model_len=MAX_MODEL_LEN) as vllm_model:
output = vllm_model.encode("Write a short story about a robot that"
" dreams for the first time.\n")
model_config = vllm_model.model.llm_engine.model_config
model_tokenizer = vllm_model.model.llm_engine.tokenizer
# asserts on the bert model config file
assert model_config.encoder_config["max_seq_length"] == 512
assert not model_config.encoder_config["do_lower_case"]
# asserts on the pooling config files
assert model_config.pooler_config.pooling_type == PoolingType.MEAN.name
> assert model_config.pooler_config.pooling_norm
E AttributeError: 'PoolerConfig' object has no attribute 'pooling_norm'. Did you mean: 'pooling_type'?
tests/model_executor/test_model_load_with_params.py:83: AttributeError
------------------------------------------------- Captured stdout call -------------------------------------------------
INFO 05-27 12:32:21 [config.py:577] Found sentence-transformers tokenize configuration.
INFO 05-27 12:32:21 [config.py:3131] Downcasting torch.float32 to torch.float16.
INFO 05-27 12:32:21 [config.py:473] Found sentence-transformers modules configuration.
INFO 05-27 12:32:21 [config.py:493] Found pooling configuration.
INFO 05-27 12:32:21 [config.py:793] This model supports multiple tasks: {'reward', 'embed', 'classify', 'score'}. Defaulting to 'embed'.
WARNING 05-27 12:32:21 [arg_utils.py:1583] --task embed is not supported by the V1 Engine. Falling back to V0.
INFO 05-27 12:32:21 [llm_engine.py:230] Initializing a V0 LLM engine (v0.9.1.dev148+gfba064270.d20250526) with config: model='intfloat/multilingual-e5-small', speculative_config=None, tokenizer='intfloat/multilingual-e5-small', skip_tokenizer_init=False, tokenizer_mode=auto, revision=main, override_neuron_config={}, tokenizer_revision=main, trust_remote_code=True, dtype=torch.float16, max_seq_len=128, download_dir=None, load_format=LoadFormat.AUTO, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto, device_config=cuda, decoding_config=DecodingConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_backend=''), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None), seed=0, served_model_name=intfloat/multilingual-e5-small, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=None, chunked_prefill_enabled=False, use_async_output_proc=False, pooler_config=PoolerConfig(pooling_type='MEAN', normalize=True, softmax=None, step_tag_id=None, returned_token_ids=None), compilation_config={"compile_sizes": [], "inductor_compile_config": {"enable_auto_functionalized_v2": false}, "cudagraph_capture_sizes": [256, 248, 240, 232, 224, 216, 208, 200, 192, 184, 176, 168, 160, 152, 144, 136, 128, 120, 112, 104, 96, 88, 80, 72, 64, 56, 48, 40, 32, 24, 16, 8, 4, 2, 1], "max_capture_size": 256}, use_cached_outputs=False,
INFO 05-27 12:32:23 [cuda.py:292] Using Flash Attention backend.
INFO 05-27 12:32:23 [parallel_state.py:1064] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, TP rank 0, EP rank 0
INFO 05-27 12:32:23 [model_runner.py:1170] Starting to load model intfloat/multilingual-e5-small...
INFO 05-27 12:32:24 [weight_utils.py:291] Using model weights format ['*.safetensors']
INFO 05-27 12:32:24 [weight_utils.py:344] No model.safetensors.index.json found in remote.
INFO 05-27 12:32:24 [default_loader.py:280] Loading weights took 0.06 seconds
INFO 05-27 12:32:24 [model_runner.py:1202] Model loading took 0.2204 GiB and 1.087685 seconds
------------------------------------------------- Captured stderr call -------------------------------------------------
Loading safetensors checkpoint shards: 0% Completed | 0/1 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:00<00:00, 18.67it/s]
Adding requests: 100%|██████████| 66/66 [00:00<00:00, 11180.74it/s]
Processed prompts: 100%|██████████| 66/66 [00:00<00:00, 2941.53it/s, est. speed input: 0.00 toks/s, output: 0.00 toks/s]
=============================================== short test summary info ================================================
FAILED tests/model_executor/test_model_load_with_params.py::test_model_loading_with_params - AttributeError: 'PoolerConfig' object has no attribute 'pooling_norm'. Did you mean: 'pooling_type'?
FAILED tests/model_executor/test_model_load_with_params.py::test_roberta_model_loading_with_params - AttributeError: 'PoolerConfig' object has no attribute 'pooling_norm'. Did you mean: 'pooling_type'?
======================================== 2 failed, 1 passed in 64.00s (0:01:04) ========================================
Before submitting a new issue...
- Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.