[Bug]: LLama4 Not working on PP #16385
Description
Your current environment
Your output of `python collect_env.py` here
🐛 Describe the bug
I am running into an issue where i am unable to launch Llama-4-Maverick-17B-128E-Instruct-FP8 in a distributed fashion using Ray.
As you can see below, VLLM is able to successfully connect to the Ray cluster, however it looks like the value for architectures appears to be None on the way workers node.
Looking through the stack trace i can see that architectures is being set to None despite both the config.json and the --hf-overides flag both specifying {"architectures": ["Llama4ForConditionalGeneration"]}
I can confirm this is only happening for llama4 and was able to successfully distribute 3.3 over 16 X A 100.
VLLM_DISABLE_COMPILE_CACHE=1 python -m vllm.entrypoints.openai.api_server --model /home/jovyan/llama4-llm-vol/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8 --served-model-name Llama-4-Maverick-17B-128E-Instruct-FP8 --enforce-eager --max-model-len 2000 --tensor-parallel 8 --pipeline-parallel-size 2 --gpu-memory-utilization 0.95 --host 0.0.0.0 --distributed-executor-backend ray --port 8000 --quantization compressed-tensors --hf-overrides '{"architectures": ["Llama4ForConditionalGeneration"]}'
INFO 04-10 03:30:20 [__init__.py:239] Automatically detected platform cuda.
INFO 04-10 03:30:22 [api_server.py:1034] vLLM API server version 0.8.3rc2.dev80+gcb84e45a
INFO 04-10 03:30:22 [api_server.py:1035] args: Namespace(host='0.0.0.0', port=8000, uvicorn_log_level='info', disable_uvicorn_access_log=False, allow_credentials=False, allowed_origins=['*'], allowed_methods=['*'], allowed_headers=['*'], api_key=None, lora_modules=None, prompt_adapters=None, chat_template=None, chat_template_content_format='auto', response_role='assistant', ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, enable_ssl_refresh=False, ssl_cert_reqs=0, root_path=None, middleware=[], return_tokens_as_token_ids=False, disable_frontend_multiprocessing=False, enable_request_id_headers=False, enable_auto_tool_choice=False, tool_call_parser=None, tool_parser_plugin='', model='/home/jovyan/llama4-llm-vol/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8', task='auto', tokenizer=None, hf_config_path=None, skip_tokenizer_init=False, revision=None, code_revision=None, tokenizer_revision=None, tokenizer_mode='auto', trust_remote_code=False, allowed_local_media_path=None, download_dir=None, load_format='auto', config_format=<ConfigFormat.AUTO: 'auto'>, dtype='auto', kv_cache_dtype='auto', max_model_len=2000, guided_decoding_backend='xgrammar', logits_processor_pattern=None, model_impl='auto', distributed_executor_backend='ray', pipeline_parallel_size=2, tensor_parallel_size=8, data_parallel_size=1, enable_expert_parallel=False, max_parallel_loading_workers=None, ray_workers_use_nsight=False, block_size=None, enable_prefix_caching=None, prefix_caching_hash_algo='builtin', disable_sliding_window=False, use_v2_block_manager=True, num_lookahead_slots=0, seed=None, swap_space=4, cpu_offload_gb=0, gpu_memory_utilization=0.95, num_gpu_blocks_override=None, max_num_batched_tokens=None, max_num_partial_prefills=1, max_long_partial_prefills=1, long_prefill_token_threshold=0, max_num_seqs=None, max_logprobs=20, disable_log_stats=False, quantization='compressed-tensors', rope_scaling=None, rope_theta=None, hf_token=None, hf_overrides={'architectures': ['Llama4ForConditionalGeneration']}, enforce_eager=True, max_seq_len_to_capture=8192, disable_custom_all_reduce=False, tokenizer_pool_size=0, tokenizer_pool_type='ray', tokenizer_pool_extra_config=None, limit_mm_per_prompt=None, mm_processor_kwargs=None, disable_mm_preprocessor_cache=False, enable_lora=False, enable_lora_bias=False, max_loras=1, max_lora_rank=16, lora_extra_vocab_size=256, lora_dtype='auto', long_lora_scaling_factors=None, max_cpu_loras=None, fully_sharded_loras=False, enable_prompt_adapter=False, max_prompt_adapters=1, max_prompt_adapter_token=0, device='auto', num_scheduler_steps=1, use_tqdm_on_load=True, multi_step_stream_outputs=True, scheduler_delay_factor=0.0, enable_chunked_prefill=None, speculative_config=None, model_loader_extra_config=None, ignore_patterns=[], preemption_mode=None, served_model_name=['Llama-4-Maverick-17B-128E-Instruct-FP8'], qlora_adapter_name_or_path=None, show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, disable_async_output_proc=False, scheduling_policy='fcfs', scheduler_cls='vllm.core.scheduler.Scheduler', override_neuron_config=None, override_pooler_config=None, compilation_config=None, kv_transfer_config=None, worker_cls='auto', worker_extension_cls='', generation_config='auto', override_generation_config=None, enable_sleep_mode=False, calculate_kv_scales=False, additional_config=None, enable_reasoning=False, reasoning_parser=None, disable_cascade_attn=False, disable_chunked_mm_input=False, disable_log_requests=False, max_log_len=None, disable_fastapi_docs=False, enable_prompt_tokens_details=False, enable_server_load_tracking=False)
INFO 04-10 03:30:22 [config.py:352] Overriding HF config with {'architectures': ['Llama4ForConditionalGeneration']}
INFO 04-10 03:30:30 [config.py:604] This model supports multiple tasks: {'score', 'reward', 'embed', 'generate', 'classify'}. Defaulting to 'generate'.
INFO 04-10 03:30:31 [config.py:1797] Chunked prefill is enabled with max_num_batched_tokens=2048.
INFO 04-10 03:30:37 [__init__.py:239] Automatically detected platform cuda.
INFO 04-10 03:30:40 [core.py:61] Initializing a V1 LLM engine (v0.8.3rc2.dev80+gcb84e45a) with config: model='/home/jovyan/llama4-llm-vol/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8', speculative_config=None, tokenizer='/home/jovyan/llama4-llm-vol/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, override_neuron_config=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=2000, download_dir=None, load_format=auto, tensor_parallel_size=8, pipeline_parallel_size=2, disable_custom_all_reduce=False, quantization=compressed-tensors, enforce_eager=True, kv_cache_dtype=auto, device_config=cuda, decoding_config=DecodingConfig(guided_decoding_backend='xgrammar', reasoning_backend=None), observability_config=ObservabilityConfig(show_hidden_metrics=False, otlp_traces_endpoint=None, collect_model_forward_time=False, collect_model_execute_time=False), seed=None, served_model_name=Llama-4-Maverick-17B-128E-Instruct-FP8, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=True, chunked_prefill_enabled=True, use_async_output_proc=False, disable_mm_preprocessor_cache=False, mm_processor_kwargs=None, pooler_config=None, compilation_config={"splitting_ops":[],"compile_sizes":[],"cudagraph_capture_sizes":[],"max_capture_size":0}
2025-04-10 03:30:40,246 INFO worker.py:1654 -- Connecting to existing Ray cluster at address: ray-head.rapid-experimentation-team.svc.cluster.local:6379...
2025-04-10 03:30:40,464 INFO worker.py:1841 -- Connected to Ray cluster.
INFO 04-10 03:30:42 [ray_utils.py:335] No current placement group found. Creating a new placement group.
INFO 04-10 03:30:42 [ray_distributed_executor.py:176] use_ray_spmd_worker: True
(pid=21440) INFO 04-10 03:30:46 [__init__.py:239] Automatically detected platform cuda.
INFO 04-10 03:30:50 [ray_distributed_executor.py:352] non_carry_over_env_vars from config: set()
INFO 04-10 03:30:50 [ray_distributed_executor.py:354] Copying the following environment variables to workers: ['LD_LIBRARY_PATH', 'VLLM_USE_RAY_SPMD_WORKER', 'VLLM_USE_RAY_COMPILED_DAG', 'VLLM_WORKER_MULTIPROC_METHOD', 'VLLM_USE_V1', 'VLLM_DISABLE_COMPILE_CACHE']
INFO 04-10 03:30:50 [ray_distributed_executor.py:357] If certain env vars should NOT be copied to workers, add them to /app/.config/vllm/ray_non_carry_over_env_vars.json file
(RayWorkerWrapper pid=21428) WARNING 04-10 03:30:50 [utils.py:2429] Methods determine_num_available_blocks,device_config,get_cache_block_size_bytes,initialize_cache not implemented in <vllm.v1.worker.gpu_worker.Worker object at 0x7f5b88740580>
(RayWorkerWrapper pid=21428) INFO 04-10 03:30:56 [utils.py:990] Found nccl from library libnccl.so.2
(RayWorkerWrapper pid=21428) INFO 04-10 03:30:56 [pynccl.py:69] vLLM is using nccl==2.21.5
(pid=16887, ip=198.18.77.250) INFO 04-10 03:30:46 [__init__.py:239] Automatically detected platform cuda. [repeated 15x across cluster] (Ray deduplicates logs by default. Set RAY_DEDUP_LOGS=0 to disable log deduplication, or see https://docs.ray.io/en/master/ray-observability/user-guides/configure-logging.html#log-deduplication for more options.)
(RayWorkerWrapper pid=16881, ip=198.18.77.250) WARNING 04-10 03:30:50 [utils.py:2429] Methods determine_num_available_blocks,device_config,get_cache_block_size_bytes,initialize_cache not implemented in <vllm.v1.worker.gpu_worker.Worker object at 0x7f44fc50a170> [repeated 15x across cluster]
(RayWorkerWrapper pid=21428) INFO 04-10 03:30:59 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /app/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3,4,5,6,7.json
(RayWorkerWrapper pid=16907, ip=198.18.77.250) INFO 04-10 03:30:59 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[1, 2, 3, 4, 5, 6, 7], buffer_handle=(7, 4194304, 6, 'psm_b575475e'), local_subscribe_addr='ipc:///tmp/b8d239f0-862f-48b9-9ee8-325b401f4b0e', remote_subscribe_addr=None, remote_addr_ipv6=False)
(RayWorkerWrapper pid=21430) INFO 04-10 03:30:59 [parallel_state.py:957] rank 1 in world size 16 is assigned as DP rank 0, PP rank 0, TP rank 1
(RayWorkerWrapper pid=21430) INFO 04-10 03:30:59 [cuda.py:221] Using Flash Attention backend on V1 engine.
(RayWorkerWrapper pid=16907, ip=198.18.77.250) INFO 04-10 03:31:04 [gpu_model_runner.py:1277] Starting to load model /home/jovyan/llama4-llm-vol/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8...
(RayWorkerWrapper pid=16881, ip=198.18.77.250) INFO 04-10 03:30:59 [utils.py:990] Found nccl from library libnccl.so.2 [repeated 31x across cluster]
(RayWorkerWrapper pid=16881, ip=198.18.77.250) INFO 04-10 03:30:59 [pynccl.py:69] vLLM is using nccl==2.21.5 [repeated 31x across cluster]
(RayWorkerWrapper pid=16881, ip=198.18.77.250) INFO 04-10 03:30:59 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /app/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3,4,5,6,7.json [repeated 15x across cluster]
(RayWorkerWrapper pid=21428) INFO 04-10 03:30:59 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[1, 2, 3, 4, 5, 6, 7], buffer_handle=(7, 4194304, 6, 'psm_215d0926'), local_subscribe_addr='ipc:///tmp/f4217ffd-9b42-480f-acf0-7b78e2d42805', remote_subscribe_addr=None, remote_addr_ipv6=False)
(RayWorkerWrapper pid=21442) INFO 04-10 03:30:59 [parallel_state.py:957] rank 5 in world size 16 is assigned as DP rank 0, PP rank 0, TP rank 5 [repeated 15x across cluster]
(RayWorkerWrapper pid=21442) INFO 04-10 03:30:59 [cuda.py:221] Using Flash Attention backend on V1 engine. [repeated 15x across cluster]
(RayWorkerWrapper pid=16907, ip=198.18.77.250) WARNING 04-10 03:31:05 [registry.py:432] No model architectures are specified
ERROR 04-10 03:31:05 [core.py:386] EngineCore hit an exception: Traceback (most recent call last):
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/v1/engine/core.py", line 377, in run_engine_core
ERROR 04-10 03:31:05 [core.py:386] engine_core = EngineCoreProc(*args, **kwargs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/v1/engine/core.py", line 319, in __init__
ERROR 04-10 03:31:05 [core.py:386] super().__init__(vllm_config, executor_class, log_stats)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/v1/engine/core.py", line 67, in __init__
ERROR 04-10 03:31:05 [core.py:386] self.model_executor = executor_class(vllm_config)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/executor/executor_base.py", line 286, in __init__
ERROR 04-10 03:31:05 [core.py:386] super().__init__(*args, **kwargs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/executor/executor_base.py", line 52, in __init__
ERROR 04-10 03:31:05 [core.py:386] self._init_executor()
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/executor/ray_distributed_executor.py", line 114, in _init_executor
ERROR 04-10 03:31:05 [core.py:386] self._init_workers_ray(placement_group)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/executor/ray_distributed_executor.py", line 396, in _init_workers_ray
ERROR 04-10 03:31:05 [core.py:386] self._run_workers("load_model",
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/executor/ray_distributed_executor.py", line 521, in _run_workers
ERROR 04-10 03:31:05 [core.py:386] ray_worker_outputs = ray.get(ray_worker_outputs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/ray/_private/auto_init_hook.py", line 21, in auto_init_wrapper
ERROR 04-10 03:31:05 [core.py:386] return fn(*args, **kwargs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/ray/_private/client_mode_hook.py", line 103, in wrapper
ERROR 04-10 03:31:05 [core.py:386] return func(*args, **kwargs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/ray/_private/worker.py", line 2771, in get
ERROR 04-10 03:31:05 [core.py:386] values, debugger_breakpoint = worker.get_objects(object_refs, timeout=timeout)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/ray/_private/worker.py", line 919, in get_objects
ERROR 04-10 03:31:05 [core.py:386] raise value.as_instanceof_cause()
ERROR 04-10 03:31:05 [core.py:386] ray.exceptions.RayTaskError(TypeError): ray::RayWorkerWrapper.execute_method() (pid=16911, ip=198.18.77.250, actor_id=c3542595396344fb7bc7e80106000000, repr=<vllm.executor.ray_utils.RayWorkerWrapper object at 0x7f98e1307a30>)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/worker/worker_base.py", line 621, in execute_method
ERROR 04-10 03:31:05 [core.py:386] raise e
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/worker/worker_base.py", line 612, in execute_method
ERROR 04-10 03:31:05 [core.py:386] return run_method(self, method, args, kwargs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/utils.py", line 2363, in run_method
ERROR 04-10 03:31:05 [core.py:386] return func(*args, **kwargs)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/v1/worker/gpu_worker.py", line 136, in load_model
ERROR 04-10 03:31:05 [core.py:386] self.model_runner.load_model()
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/v1/worker/gpu_model_runner.py", line 1280, in load_model
ERROR 04-10 03:31:05 [core.py:386] self.model = get_model(vllm_config=self.vllm_config)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/model_loader/__init__.py", line 14, in get_model
ERROR 04-10 03:31:05 [core.py:386] return loader.load_model(vllm_config=vllm_config)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/model_loader/loader.py", line 452, in load_model
ERROR 04-10 03:31:05 [core.py:386] model = _initialize_model(vllm_config=vllm_config)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/model_loader/loader.py", line 133, in _initialize_model
ERROR 04-10 03:31:05 [core.py:386] return model_class(vllm_config=vllm_config, prefix=prefix)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/models/mllama4.py", line 692, in __init__
ERROR 04-10 03:31:05 [core.py:386] vllm_config=vllm_config.with_hf_config(config.text_config),
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/config.py", line 3568, in with_hf_config
ERROR 04-10 03:31:05 [core.py:386] return replace(self, model_config=model_config)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/dataclasses.py", line 1453, in replace
ERROR 04-10 03:31:05 [core.py:386] return obj.__class__(**changes)
ERROR 04-10 03:31:05 [core.py:386] File "<string>", line 19, in __init__
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/config.py", line 3577, in __post_init__
ERROR 04-10 03:31:05 [core.py:386] self.model_config.verify_with_parallel_config(self.parallel_config)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/config.py", line 797, in verify_with_parallel_config
ERROR 04-10 03:31:05 [core.py:386] if not self.registry.is_pp_supported_model(self.architectures):
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/models/registry.py", line 501, in is_pp_supported_model
ERROR 04-10 03:31:05 [core.py:386] model_cls, _ = self.inspect_model_cls(architectures)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/models/registry.py", line 447, in inspect_model_cls
ERROR 04-10 03:31:05 [core.py:386] architectures = self._normalize_archs(architectures)
ERROR 04-10 03:31:05 [core.py:386] File "/opt/conda/lib/python3.10/site-packages/vllm/model_executor/models/registry.py", line 436, in _normalize_archs
ERROR 04-10 03:31:05 [core.py:386] filter(lambda model: model in self.models, architectures))
ERROR 04-10 03:31:05 [core.py:386] TypeError: 'NoneType' object is not iterable
ERROR 04-10 03:31:05 [core.py:386]
INFO 04-10 03:31:05 [ray_distributed_executor.py:127] Shutting down Ray distributed executor. If you see error log from logging.cc regarding SIGTERM received, please ignore because this is the expected termination process in Ray.
CRITICAL 04-10 03:31:05 [core_client.py:359] Got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
ENVIRONMENT
PyTorch version: 2.6.0+cu124
Is debug build: False
CUDA used to build PyTorch: 12.4
ROCM used to build PyTorch: N/A
OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: Could not collect
Libc version: glibc-2.35
Python version: 3.10.16 | packaged by conda-forge | (main, Dec 5 2024, 14:16:10) [GCC 13.3.0] (64-bit runtime)
Python platform: Linux-5.10.234-225.895.amzn2.x86_64-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.1.105
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration:
GPU 0: NVIDIA A100-SXM4-40GB
GPU 1: NVIDIA A100-SXM4-40GB
GPU 2: NVIDIA A100-SXM4-40GB
GPU 3: NVIDIA A100-SXM4-40GB
GPU 4: NVIDIA A100-SXM4-40GB
GPU 5: NVIDIA A100-SXM4-40GB
GPU 6: NVIDIA A100-SXM4-40GB
GPU 7: NVIDIA A100-SXM4-40GB
Nvidia driver version: 550.144.03
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.0
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 46 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 96
On-line CPU(s) list: 0-95
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Platinum 8275CL CPU @ 3.00GHz
CPU family: 6
Model: 85
Thread(s) per core: 2
Core(s) per socket: 24
Socket(s): 2
Stepping: 7
BogoMIPS: 5999.98
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 1.5 MiB (48 instances)
L1i cache: 1.5 MiB (48 instances)
L2 cache: 48 MiB (48 instances)
L3 cache: 71.5 MiB (2 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0-23,48-71
NUMA node1 CPU(s): 24-47,72-95
Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
Vulnerability Itlb multihit: KVM: Mitigation: VMX unsupported
Vulnerability L1tf: Mitigation; PTE Inversion
Vulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Meltdown: Mitigation; PTI
Vulnerability Mmio stale data: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Vulnerable
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Vulnerable
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] nvidia-cublas-cu12==12.4.5.8
[pip3] nvidia-cuda-cupti-cu12==12.4.127
[pip3] nvidia-cuda-nvrtc-cu12==12.4.127
[pip3] nvidia-cuda-runtime-cu12==12.4.127
[pip3] nvidia-cudnn-cu12==9.1.0.70
[pip3] nvidia-cufft-cu12==11.2.1.3
[pip3] nvidia-curand-cu12==10.3.5.147
[pip3] nvidia-cusolver-cu12==11.6.1.9
[pip3] nvidia-cusparse-cu12==12.3.1.170
[pip3] nvidia-cusparselt-cu12==0.6.2
[pip3] nvidia-nccl-cu12==2.21.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] pytorch-lightning==2.5.0.post0
[pip3] pyzmq==26.4.0
[pip3] torch==2.6.0
[pip3] torchaudio==2.6.0
[pip3] torchmetrics==1.6.3
[pip3] torchvision==0.21.0
[pip3] transformers==4.51.1
[pip3] triton==3.2.0
[conda] numpy 1.26.4 pypi_0 pypi
[conda] nvidia-cublas-cu12 12.4.5.8 pypi_0 pypi
[conda] nvidia-cuda-cupti-cu12 12.4.127 pypi_0 pypi
[conda] nvidia-cuda-nvrtc-cu12 12.4.127 pypi_0 pypi
[conda] nvidia-cuda-runtime-cu12 12.4.127 pypi_0 pypi
[conda] nvidia-cudnn-cu12 9.1.0.70 pypi_0 pypi
[conda] nvidia-cufft-cu12 11.2.1.3 pypi_0 pypi
[conda] nvidia-curand-cu12 10.3.5.147 pypi_0 pypi
[conda] nvidia-cusolver-cu12 11.6.1.9 pypi_0 pypi
[conda] nvidia-cusparse-cu12 12.3.1.170 pypi_0 pypi
[conda] nvidia-cusparselt-cu12 0.6.2 pypi_0 pypi
[conda] nvidia-nccl-cu12 2.21.5 pypi_0 pypi
[conda] nvidia-nvjitlink-cu12 12.4.127 pypi_0 pypi
[conda] nvidia-nvtx-cu12 12.4.127 pypi_0 pypi
[conda] pytorch-lightning 2.5.0.post0 pypi_0 pypi
[conda] pyzmq 26.4.0 pypi_0 pypi
[conda] torch 2.6.0 pypi_0 pypi
[conda] torchaudio 2.6.0 pypi_0 pypi
[conda] torchmetrics 1.6.3 pypi_0 pypi
[conda] torchvision 0.21.0 pypi_0 pypi
[conda] transformers 4.51.1 pypi_0 pypi
[conda] triton 3.2.0 pypi_0 pypi
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.