[Bug]: `RuntimeError: Failed to infer device type` with v0.7.2

[imangohari1]

Your current environment

The output of `python collect_env.py`

INFO 02-06 18:50:45 __init__.py:194] No platform detected, vLLM is running on UnspecifiedPlatform
Collecting environment information...
PyTorch version: 2.5.1+cu124
Is debug build: False
CUDA used to build PyTorch: 12.4
ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.1 LTS (x86_64)
GCC version: (Ubuntu 13.2.0-23ubuntu4) 13.2.0
Clang version: Could not collect
CMake version: version 3.31.1
Libc version: glibc-2.39

Python version: 3.12.3 (main, Nov  6 2024, 18:32:19) [GCC 13.2.0] (64-bit runtime)
Python platform: Linux-6.8.0-52-generic-x86_64-with-glibc2.39
Is CUDA available: True
CUDA runtime version: 12.6.85
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA A100 80GB PCIe
Nvidia driver version: 565.57.01
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.6.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.6.0
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:                        52 bits physical, 57 bits virtual
Byte Order:                           Little Endian
CPU(s):                               224
On-line CPU(s) list:                  0-223
Vendor ID:                            GenuineIntel
BIOS Vendor ID:                       Intel(R) Corporation
Model name:                           Intel(R) Xeon(R) Platinum 8480+
BIOS Model name:                      Intel(R) Xeon(R) Platinum 8480+  CPU @ 2.0GHz
BIOS CPU family:                      179
CPU family:                           6
Model:                                143
Thread(s) per core:                   2
Core(s) per socket:                   56
Socket(s):                            2
Stepping:                             6
CPU(s) scaling MHz:                   25%
CPU max MHz:                          3800.0000
CPU min MHz:                          800.0000
BogoMIPS:                             4000.00
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cat_l2 cdp_l3 intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect user_shstk avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid bus_lock_detect cldemote movdiri movdir64b enqcmd fsrm md_clear serialize tsxldtrk pconfig arch_lbr ibt amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
Virtualization:                       VT-x
L1d cache:                            5.3 MiB (112 instances)
L1i cache:                            3.5 MiB (112 instances)
L2 cache:                             224 MiB (112 instances)
L3 cache:                             210 MiB (2 instances)
NUMA node(s):                         2
NUMA node0 CPU(s):                    0-55,112-167
NUMA node1 CPU(s):                    56-111,168-223
Vulnerability Gather data sampling:   Not affected
Vulnerability Itlb multihit:          Not affected
Vulnerability L1tf:                   Not affected
Vulnerability Mds:                    Not affected
Vulnerability Meltdown:               Not affected
Vulnerability Mmio stale data:        Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed:               Not affected
Vulnerability Spec rstack overflow:   Not affected
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:             Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] mypy-extensions==1.0.0
[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-cudnn-frontend==1.8.0
[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-dali-cuda120==1.44.0
[pip3] nvidia-ml-py==12.570.86
[pip3] nvidia-modelopt==0.21.0
[pip3] nvidia-nccl-cu12==2.21.5
[pip3] nvidia-nvimgcodec-cu12==0.3.0.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] nvidia-pyindex==1.0.9
[pip3] onnx==1.17.0
[pip3] optree==0.13.1
[pip3] pynvml==11.4.1
[pip3] pytorch-triton==3.0.0+72734f086
[pip3] pyzmq==26.2.0
[pip3] torch==2.5.1
[pip3] torch_tensorrt==2.6.0a0
[pip3] torchaudio==2.5.1
[pip3] torchprofile==0.0.4
[pip3] torchvision==0.20.1
[pip3] transformers==4.48.2
[pip3] triton==3.1.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.7.3.dev10+g467a96a5
vLLM Build Flags:
CUDA Archs: 7.0 7.5 8.0 8.6 9.0+PTX; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      0-55,112-167    0               N/A

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks

NVIDIA_VISIBLE_DEVICES=all
CUBLAS_VERSION=12.6.4.1
NVIDIA_REQUIRE_CUDA=cuda>=9.0
CUDA_CACHE_DISABLE=1
TORCH_CUDA_ARCH_LIST=7.0 7.5 8.0 8.6 9.0+PTX
NCCL_VERSION=2.23.4
NVIDIA_DRIVER_CAPABILITIES=compute,utility,video
NVIDIA_PRODUCT_NAME=PyTorch
CUDA_VERSION=12.6.3.004
PYTORCH_VERSION=2.6.0a0+df5bbc0
PYTORCH_BUILD_NUMBER=0
CUDNN_FRONTEND_VERSION=1.8.0
CUDNN_VERSION=9.6.0.74
PYTORCH_HOME=/opt/pytorch/pytorch
LD_LIBRARY_PATH=/usr/local/lib/python3.12/dist-packages/torch/lib:/usr/local/lib/python3.12/dist-packages/torch_tensorrt/lib:/usr/local/cuda/compat/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
NVIDIA_BUILD_ID=126674149
CUDA_DRIVER_VERSION=560.35.05
PYTORCH_BUILD_VERSION=2.6.0a0+df5bbc0
CUDA_HOME=/usr/local/cuda
CUDA_HOME=/usr/local/cuda
CUDA_MODULE_LOADING=LAZY
NVIDIA_REQUIRE_JETPACK_HOST_MOUNTS=
NVIDIA_PYTORCH_VERSION=24.12
TORCH_ALLOW_TF32_CUBLAS_OVERRIDE=1
NCCL_CUMEM_ENABLE=0
TORCHINDUCTOR_COMPILE_THREADS=1

🐛 Describe the bug

Using the main branch on commit 467a96a5415dc896170cecc0bb83d9c49c2f3c5e, we are facing RuntimeError: Failed to infer device type. The details of the env/test is given below.

Note

This issue does NOT happen using 0.7.1 release. On the same machine, same container, changing the installation to pip install vllm (or pip install https://github.com/vllm-project/vllm/releases/download/v0.7.1/vllm-0.7.1-cp38-abi3-manylinux1_x86_64.whl) works fine.

Container/Setup

• Container: nvcr.io/nvidia/pytorch:24.12-py3
• Setup:

git clone https://github.com/vllm-project/vllm.git
cd vllm
VLLM_USE_PRECOMPILED=1 pip install --editable .

Test

python -m vllm.entrypoints.openai.api_server --host 0.0.0.0 --port 8000 --model meta-llama/Llama-3.2-3B-Instruct --seed 42 -tp 1 --use-v2-block-manager --max_model_len 2048

INFO 02-06 19:18:33 __init__.py:194] No platform detected, vLLM is running on UnspecifiedPlatform
INFO 02-06 19:18:34 api_server.py:840] vLLM API server version 0.7.3.dev10+g467a96a5
INFO 02-06 19:18:34 api_server.py:841] args: Namespace(host='0.0.0.0', port=8000, uvicorn_log_level='info', 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, 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, enable_reasoning=False, reasoning_parser=None, tool_call_parser=None, tool_parser_plugin='', model='meta-llama/Llama-3.2-3B-Instruct', task='auto', tokenizer=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=2048, guided_decoding_backend='xgrammar', logits_processor_pattern=None, model_impl='auto', distributed_executor_backend=None, pipeline_parallel_size=1, tensor_parallel_size=1, max_parallel_loading_workers=None, ray_workers_use_nsight=False, block_size=None, enable_prefix_caching=None, disable_sliding_window=False, use_v2_block_manager=True, num_lookahead_slots=0, seed=42, swap_space=4, cpu_offload_gb=0, gpu_memory_utilization=0.9, num_gpu_blocks_override=None, max_num_batched_tokens=None, max_num_seqs=None, max_logprobs=20, disable_log_stats=False, quantization=None, rope_scaling=None, rope_theta=None, hf_overrides=None, enforce_eager=False, 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, multi_step_stream_outputs=True, scheduler_delay_factor=0.0, enable_chunked_prefill=None, speculative_model=None, speculative_model_quantization=None, num_speculative_tokens=None, speculative_disable_mqa_scorer=False, speculative_draft_tensor_parallel_size=None, speculative_max_model_len=None, speculative_disable_by_batch_size=None, ngram_prompt_lookup_max=None, ngram_prompt_lookup_min=None, spec_decoding_acceptance_method='rejection_sampler', typical_acceptance_sampler_posterior_threshold=None, typical_acceptance_sampler_posterior_alpha=None, disable_logprobs_during_spec_decoding=None, model_loader_extra_config=None, ignore_patterns=[], preemption_mode=None, served_model_name=None, qlora_adapter_name_or_path=None, otlp_traces_endpoint=None, collect_detailed_traces=None, disable_async_output_proc=False, scheduling_policy='fcfs', override_neuron_config=None, override_pooler_config=None, compilation_config=None, kv_transfer_config=None, worker_cls='auto', generation_config=None, override_generation_config=None, enable_sleep_mode=False, calculate_kv_scales=False, disable_log_requests=False, max_log_len=None, disable_fastapi_docs=False, enable_prompt_tokens_details=False)
INFO 02-06 19:18:34 api_server.py:206] Started engine process with PID 879
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/tmp/vllm/vllm/entrypoints/openai/api_server.py", line 911, in <module>
    uvloop.run(run_server(args))
  File "/usr/local/lib/python3.12/dist-packages/uvloop/__init__.py", line 109, in run
    return __asyncio.run(
           ^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 194, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "uvloop/loop.pyx", line 1518, in uvloop.loop.Loop.run_until_complete
  File "/usr/local/lib/python3.12/dist-packages/uvloop/__init__.py", line 61, in wrapper
    return await main
           ^^^^^^^^^^
  File "/tmp/vllm/vllm/entrypoints/openai/api_server.py", line 875, in run_server
    async with build_async_engine_client(args) as engine_client:
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/vllm/vllm/entrypoints/openai/api_server.py", line 136, in build_async_engine_client
    async with build_async_engine_client_from_engine_args(
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/vllm/vllm/entrypoints/openai/api_server.py", line 217, in build_async_engine_client_from_engine_args
    engine_config = engine_args.create_engine_config()
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/vllm/vllm/engine/arg_utils.py", line 1074, in create_engine_config
    device_config = DeviceConfig(device=self.device)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/vllm/vllm/config.py", line 1626, in __init__
    raise RuntimeError("Failed to infer device type")

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.

[youkaichao]

looks strange. I think #12809 should solve it. the actual error should raise earlier.

[youkaichao]

INFO 02-06 18:50:45 init.py:194] No platform detected, vLLM is running on UnspecifiedPlatform

this is the problem.

can you try to follow

vllm/vllm/utils.py

Line 2223 in b260782

def import_pynvml():

to see why it happens?

[pip3] nvidia-ml-py==12.570.86
[pip3] pynvml==11.4.1

since you have both pynvml and nvidia-ml-py, I think the code above should work. Not sure why it does not work for your case.

[lizongyao123]

升级pynvml到12.0.0即可。

pip install -U pynvml

[youkaichao]

@lizongyao123 @imangohari1 can you install the latest release (0.7.2 has been released) and see if it happens again? I think the latest release should work even if you have pynvml < 12

[ghosthamlet]

@lizongyao123 @imangohari1 can you install the latest release (0.7.2 has been released) and see if it happens again? I think the latest release should work even if you have pynvml < 12

I have same problem: No platform detected, vLLM is running on UnspecifiedPlatform
vllm 0.7.2
Run import_pynvml()
RuntimeError: You are using a deprecated pynvmlpackage. Please uninstallpynvml or upgrade to at least version 12.0. See https://pypi.org/project/pynvml for more information.

After ungrade pynvml to 12.0.0
Fix it.

[zimoqingfeng]

@lizongyao123 @imangohari1 can you install the latest release (0.7.2 has been released) and see if it happens again? I think the latest release should work even if you have pynvml < 12

It may not work when I use vllm=0.7.2 & pynvml==11.5.3

[2catycm]

I encountered similar problem

[2catycm]

Very good issue，I also succeeded to fix this by uv pip install pynvml==12.0.0

[youkaichao]

@ghosthamlet are you using vllm serve , or use vllm as a library inside a python file?

[youkaichao]

If anyone can give me a reproducible example, that would be great. In addition, feel free to join https://slack.vllm.ai for quick communication.

[imangohari1]

@youkaichao

If anyone can give me a reproducible example, that would be great. In addition, feel free to join https://slack.vllm.ai for quick communication.

I have included a reproducer in the #12847 (comment).
Have you tried this?
I am gonna try to join the slack channel. I can't join the slack channel due to internal policy restrictions :/

[imangohari1]

@lizongyao123 @imangohari1 can you install the latest release (0.7.2 has been released) and see if it happens again? I think the latest release should work even if you have pynvml < 12

@youkaichao I redid the test with 0.7.2 release with pynvml 11.4.1, and still fails.

Note

The only workaround right now is to update pynvml to 12.0.0 with pip install -U pynvml

[youkaichao]

@imangohari1 thanks! I find the root cause, the docker image nvcr.io/nvidia/pytorch:24.12-py3 includes pynvml==11.4.1