[Usage]: Trying to run VLLM on L4 machine, but got "Expected all tensors to be on the same device" exception

vllm-project / vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

Apache License 2.0

28.86k stars 4.28k forks source link

Your current environment

PyTorch version: 2.1.2+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Debian GNU/Linux 11 (bullseye) (x86_64)
GCC version: (Debian 10.2.1-6) 10.2.1 20210110
Clang version: Could not collect
CMake version: version 3.18.4
Libc version: glibc-2.31

Python version: 3.10.13 | packaged by conda-forge | (main, Dec 23 2023, 15:36:39) [GCC 12.3.0] (64-bit runtime)
Python platform: Linux-5.10.0-28-cloud-amd64-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: 12.1.105
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA L4
Nvidia driver version: 535.86.10
cuDNN version: Could not collect
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
Byte Order:                         Little Endian
Address sizes:                      46 bits physical, 48 bits virtual
CPU(s):                             4
On-line CPU(s) list:                0-3
Thread(s) per core:                 2
Core(s) per socket:                 2
Socket(s):                          1
NUMA node(s):                       1
Vendor ID:                          GenuineIntel
CPU family:                         6
Model:                              85
Model name:                         Intel(R) Xeon(R) CPU @ 2.20GHz
Stepping:                           7
CPU MHz:                            2200.226
BogoMIPS:                           4400.45
Hypervisor vendor:                  KVM
Virtualization type:                full
L1d cache:                          64 KiB
L1i cache:                          64 KiB
L2 cache:                           2 MiB
L3 cache:                           38.5 MiB
NUMA node0 CPU(s):                  0-3
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit:        Not affected
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Mitigation; Clear CPU buffers; SMT Host state unknown
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Retbleed:             Mitigation; Enhanced IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Mitigation; Clear CPU buffers; SMT Host state unknown
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 rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves arat avx512_vnni md_clear arch_capabilities

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] onnx==1.15.0
[pip3] onnxruntime==1.17.1
[pip3] open-clip-torch==2.24.0
[pip3] pytorch-lightning==2.0.7
[pip3] torch==2.1.2
[pip3] torchdiffeq==0.2.3
[pip3] torchmetrics==1.3.1
[pip3] torchsde==0.2.6
[pip3] torchvision==0.16.2
[pip3] triton==2.1.0
[conda] numpy                     1.25.2                   pypi_0    pypiROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.3.3
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      0-3             N/A             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

I am loading the model using the default method:

LLM(model=setting.model_repo, tensor_parallel_size=1, gpu_memory_utilization=0.9)

But I got this error:

python3.10/site-packages/nncf/torch/dynamic_graph/wrappers.py", line 82, in wrapped
    op1 = operator(*args, **kwargs)
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument index in method wrapper_CUDA__index_select)

But the document does not specify how to enforce all the tensor run on the same devices.

How would you like to use vllm

I want to run inference of a gemma-2b model. I don't know how to integrate it with vllm.

from vllm import LLM, SamplingParams prompts = [ "Hello, my name is", "The president of the United States is", "The capital of France is", "The future of AI is", ] sampling_params = SamplingParams(temperature=0.8, top_p=0.95) llm = LLM(model="google/gemma-2b-it", tensor_parallel_size=1, gpu_memory_utilization=0.9) outputs = llm.generate(prompts, sampling_params) # Print the outputs. for output in outputs: prompt = output.prompt generated_text = output.outputs[0].text print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

Collecting environment information... PyTorch version: 2.2.0.dev20231115+cu121 Is debug build: False CUDA used to build PyTorch: 12.1 ROCM used to build PyTorch: N/A OS: Debian GNU/Linux 11 (bullseye) (x86_64) GCC version: (Debian 10.2.1-6) 10.2.1 20210110 Clang version: Could not collect CMake version: version 3.18.4 Libc version: glibc-2.31 Python version: 3.9.18 | packaged by conda-forge | (main, Dec 23 2023, 16:33:10) [GCC 12.3.0] (64-bit runtime) Python platform: Linux-5.10.0-26-cloud-amd64-x86_64-with-glibc2.31 Is CUDA available: True CUDA runtime version: 12.1.105 CUDA_MODULE_LOADING set to: LAZY GPU models and configuration: GPU 0: NVIDIA L4 GPU 1: NVIDIA L4 Nvidia driver version: 535.86.10 cuDNN version: Could not collect 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 Byte Order: Little Endian Address sizes: 46 bits physical, 48 bits virtual CPU(s): 24 On-line CPU(s) list: 0-23 Thread(s) per core: 2 Core(s) per socket: 12 Socket(s): 1 NUMA node(s): 1 Vendor ID: GenuineIntel CPU family: 6 Model: 85 Model name: Intel(R) Xeon(R) CPU @ 2.20GHz Stepping: 7 CPU MHz: 2200.212 BogoMIPS: 4400.42 Hypervisor vendor: KVM Virtualization type: full L1d cache: 384 KiB L1i cache: 384 KiB L2 cache: 12 MiB L3 cache: 38.5 MiB NUMA node0 CPU(s): 0-23 Vulnerability Gather data sampling: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Mitigation; Clear CPU buffers; SMT Host state unknown Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown Vulnerability Retbleed: Mitigation; Enhanced IBRS Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT Host state unknown 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 rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves arat avx512_vnni md_clear arch_capabilities Versions of relevant libraries: [pip3] mypy==0.991 [pip3] mypy-extensions==1.0.0 [pip3] numpy==1.26.4 [pip3] pytorch-triton==2.1.0+6e4932cda8 [pip3] torch==2.2.0.dev20231115+cu121 [pip3] triton==2.2.0 [conda] numpy 1.26.4 pypi_0 pypi [conda] pytorch-triton 2.1.0+6e4932cda8 pypi_0 pypi [conda] torch 2.2.0.dev20231115+cu121 pypi_0 pypi [conda] triton 2.2.0 pypi_0 pypiROCM Version: Could not collect Neuron SDK Version: N/A vLLM Version: 0.3.3 vLLM Build Flags: CUDA Archs: 8.9; ROCm: Disabled; Neuron: Disabled GPU Topology: GPU0 GPU1 CPU Affinity NUMA Affinity GPU NUMA ID GPU0 X PHB 0-23 N/A N/A GPU1 PHB X 0-23 N/A 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

vllm-project / vllm

[Usage]: Trying to run VLLM on L4 machine, but got "Expected all tensors to be on the same device" exception #3482

Your current environment

How would you like to use vllm