[Bug]: CUDA error when serving MiniCPM-V model #23954
Description
Your current environment
The output of python collect_env.py
Collecting environment information...
==============================
System Info
==============================
OS : Ubuntu 24.04.2 LTS (x86_64)
GCC version : (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
Clang version : 18.1.3 (1ubuntu1)
CMake version : version 3.28.3
Libc version : glibc-2.39
==============================
PyTorch Info
==============================
PyTorch version : 2.8.0+cu128
Is debug build : False
CUDA used to build PyTorch : 12.8
ROCM used to build PyTorch : N/A
==============================
Python Environment
==============================
Python version : 3.12.3 (main, Aug 14 2025, 17:47:21) [GCC 13.3.0] (64-bit runtime)
Python platform : Linux-6.8.0-78-generic-x86_64-with-glibc2.39
==============================
CUDA / GPU Info
==============================
Is CUDA available : True
CUDA runtime version : 12.9.86
CUDA_MODULE_LOADING set to : LAZY
GPU models and configuration :
GPU 0: NVIDIA L40
GPU 1: NVIDIA L40
GPU 2: NVIDIA L40
GPU 3: NVIDIA L40
Nvidia driver version : 575.64.03
cuDNN version : Could not collect
HIP runtime version : N/A
MIOpen runtime version : N/A
Is XNNPACK available : True
==============================
CPU Info
==============================
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): 240
On-line CPU(s) list: 0-239
Vendor ID: GenuineIntel
Model name: INTEL(R) XEON(R) PLATINUM 8581C
CPU family: 6
Model: 207
Thread(s) per core: 2
Core(s) per socket: 60
Socket(s): 2
Stepping: 2
CPU(s) scaling MHz: 20%
CPU max MHz: 4000.0000
CPU min MHz: 800.0000
BogoMIPS: 4200.00
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 hfi 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: 5.6 MiB (120 instances)
L1i cache: 3.8 MiB (120 instances)
L2 cache: 240 MiB (120 instances)
L3 cache: 600 MiB (2 instances)
NUMA node(s): 4
NUMA node0 CPU(s): 0-29,120-149
NUMA node1 CPU(s): 30-59,150-179
NUMA node2 CPU(s): 60-89,180-209
NUMA node3 CPU(s): 90-119,210-239
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==1.1.0
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.13.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-cufile-cu12==1.13.1.3
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-ml-py==12.575.51
[pip3] nvidia-nccl-cu12==2.27.3
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvshmem-cu12==3.3.9
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pynvml==12.0.0
[pip3] pyzmq==27.0.1
[pip3] torch==2.8.0
[pip3] torch-tb-profiler==0.4.3
[pip3] torchaudio==2.8.0
[pip3] torchvision==0.23.0
[pip3] transformers==4.55.2
[pip3] triton==3.4.0
[conda] Could not collect
==============================
vLLM Info
==============================
ROCM Version : Could not collect
Neuron SDK Version : N/A
vLLM Version : 0.10.2.dev378+g4d7fe40fc.d20250829 (git sha: 4d7fe40fc, date: 20250829)
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0 GPU1 GPU2 GPU3 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X SYS SYS SYS 0-29,120-149 0 N/A
GPU1 SYS X SYS SYS 30-59,150-179 1 N/A
GPU2 SYS SYS X NODE 60-89,180-209 2 N/A
GPU3 SYS SYS NODE X 60-89,180-209 2 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
==============================
Environment Variables
==============================
CUDA_INSTALL_PATH=/usr/local/cuda-12.9
LD_LIBRARY_PATH=/usr/local/cuda-12.9/lib64
CUDACXX=/usr/local/cuda-12.9/bin/nvcc
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY
🐛 Describe the bug
How to reproduce
vllm serve /data/datasets/models-hf/MiniCPM-V-4_5/ --trust-remote-code --served-model-name minicpm
vllm bench serve \
--endpoint-type openai-chat \
--endpoint /v1/chat/completions \
--model minicpm \
--dataset-name random-mm \
--random-mm-base-items-per-request 3 \
--tokenizer /data/datasets/models-hf/MiniCPM-V-4_5 --trust-remote-code \
--num-prompts 70 \
--max-concurrency 10```
part of error log
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/compilation/cuda_graph.py", line 119, in call
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return self.runnable(*args, **kwargs)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/compilation/cuda_piecewise_backend.py", line 96, in call
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return self.compiled_graph_for_general_shape(*args)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_inductor/standalone_compile.py", line 62, in call
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return self._compiled_fn(*args)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py", line 929, in _fn
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return fn(*args, **kwargs)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_functorch/aot_autograd.py", line 1241, in forward
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return compiled_fn(full_args)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 384, in runtime_wrapper
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] all_outs = call_func_at_runtime_with_args(
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/utils.py", line 126, in call_func_at_runtime_with_args
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] out = normalize_as_list(f(args))
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 750, in inner_fn
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] outs = compiled_fn(args)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 556, in wrapper
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return compiled_fn(runtime_args)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_inductor/output_code.py", line 584, in call
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return self.current_callable(inputs)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_inductor/utils.py", line 2716, in run
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] out = model(new_inputs)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/tmp/torchinductor_zjy/sg/csgavmys24cmbt6oddae4zkyixhdv4kajshz4qm446s7mixidxsz.py", line 950, in call
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] triton_poi_fused_cat_4.run(buf12, arg10_1, arg11_1, buf14, triton_poi_fused_cat_4_xnumel, stream=stream1)
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_inductor/runtime/triton_heuristics.py", line 1180, in run
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return launcher(
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "", line 5, in launcher
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/_inductor/runtime/static_cuda_launcher.py", line 227, in run
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] _StaticCudaLauncher._launch_kernel(
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611] RuntimeError: CUDA driver error: device-side assert triggered
(VllmWorker TP1 pid=1035574) ERROR 08-30 01:28:29 [multiproc_executor.py:611]
/pytorch/aten/src/ATen/native/cuda/IndexKernel.cu:402: masked_scatter_size_check: block: [0,0,0], thread: [0,0,0] Assertion totalElements <= srcSize failed.
/pytorch/aten/src/ATen/native/cuda/IndexKernel.cu:402: masked_scatter_size_check: block: [0,0,0], thread: [0,0,0] Assertion totalElements <= srcSize failed.
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] WorkerProc hit an exception.
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] Traceback (most recent call last):
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/executor/multiproc_executor.py", line 606, in worker_busy_loop
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] output = func(*args, **kwargs)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return func(*args, **kwargs)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/worker/gpu_worker.py", line 363, in execute_model
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] output = self.model_runner.execute_model(scheduler_output,
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return func(*args, **kwargs)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/worker/gpu_model_runner.py", line 1697, in execute_model
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] valid_sampled_token_ids = self._to_list(sampled_token_ids)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/worker/gpu_model_runner.py", line 3271, in _to_list
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] self.transfer_event.synchronize()
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/cuda/streams.py", line 231, in synchronize
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] super().synchronize()
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] torch.AcceleratorError: CUDA error: device-side assert triggered
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] For debugging consider passing CUDA_LAUNCH_BLOCKING=1
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611]
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] Traceback (most recent call last):
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/executor/multiproc_executor.py", line 606, in worker_busy_loop
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] output = func(*args, **kwargs)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return func(*args, **kwargs)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/worker/gpu_worker.py", line 363, in execute_model
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] output = self.model_runner.execute_model(scheduler_output,
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] return func(*args, **kwargs)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/worker/gpu_model_runner.py", line 1697, in execute_model
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] valid_sampled_token_ids = self._to_list(sampled_token_ids)
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/vllm/v1/worker/gpu_model_runner.py", line 3271, in _to_list
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] self.transfer_event.synchronize()
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] File "/home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/cuda/streams.py", line 231, in synchronize
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] super().synchronize()
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] torch.AcceleratorError: CUDA error: device-side assert triggered
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] For debugging consider passing CUDA_LAUNCH_BLOCKING=1
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611] Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611]
(VllmWorker TP0 pid=1035573) ERROR 08-30 01:28:29 [multiproc_executor.py:611]
(EngineCore_0 pid=1035378) ERROR 08-30 01:28:29 [dump_input.py:69] Dumping input data for V1 LLM engine (v0.10.2.dev378+g4d7fe40fc.d20250829) with config: model='./hf/MiniCPM-V-2_6/', speculative_config=None, tokenizer='./hf/MiniCPM-V-2_6/', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, override_neuron_config={}, tokenizer_revision=None, trust_remote_code=True, dtype=torch.bfloat16, max_seq_len=32768, download_dir=None, load_format=auto, tensor_parallel_size=2, 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=./hf/MiniCPM-V-2_6/, enable_prefix_caching=True, chunked_prefill_enabled=True, use_async_output_proc=True, pooler_config=None, compilation_config={"level":3,"debug_dump_path":"","cache_dir":"","backend":"","custom_ops":[],"splitting_ops":["vllm.unified_attention","vllm.unified_attention_with_output","vllm.mamba_mixer2","vllm.mamba_mixer","vllm.short_conv","vllm.linear_attention"],"use_inductor":true,"compile_sizes":[],"inductor_compile_config":{"enable_auto_functionalized_v2":false},"inductor_passes":{},"cudagraph_mode":1,"use_cudagraph":true,"cudagraph_num_of_warmups":1,"cudagraph_capture_sizes":[512,504,496,488,480,472,464,456,448,440,432,424,416,408,400,392,384,376,368,360,352,344,336,328,320,312,304,296,288,280,272,264,256,248,240,232,224,216,208,200,192,184,176,168,160,152,144,136,128,120,112,104,96,88,80,72,64,56,48,40,32,24,16,8,4,2,1],"cudagraph_copy_inputs":false,"full_cuda_graph":false,"pass_config":{},"max_capture_size":512,"local_cache_dir":null},
[rank0]:[E830 01:28:29.045042152 ProcessGroupNCCL.cpp:2068] [PG ID 2 PG GUID 3 Rank 0] Process group watchdog thread terminated with exception: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
Exception raised from c10_cuda_check_implementation at /pytorch/c10/cuda/CUDAException.cpp:42 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits, std::allocator >) + 0x80 (0x7f6ab4f7eeb0 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libc10.so)
frame #1: + 0x111c7 (0x7f6ab53a51c7 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libc10_cuda.so)
frame #2: c10d::ProcessGroupNCCL::WorkNCCL::finishedGPUExecutionInternal() const + 0x50 (0x7f6a58030640 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #3: c10d::ProcessGroupNCCL::WorkNCCL::isCompleted() + 0x68 (0x7f6a5803fe28 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #4: c10d::ProcessGroupNCCL::Watchdog::runLoop() + 0x978 (0x7f6a58042f48 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #5: c10d::ProcessGroupNCCL::Watchdog::run() + 0xd2 (0x7f6a58044ec2 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #6: + 0xecdb4 (0x7f6a3b4ecdb4 in /lib/x86_64-linux-gnu/libstdc++.so.6)
frame #7: + 0x9caa4 (0x7f6ab5e9caa4 in /lib/x86_64-linux-gnu/libc.so.6)
frame #8: + 0x129c3c (0x7f6ab5f29c3c in /lib/x86_64-linux-gnu/libc.so.6)
terminate called after throwing an instance of 'c10::DistBackendError'
what(): [PG ID 2 PG GUID 3 Rank 0] Process group watchdog thread terminated with exception: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
Exception raised from c10_cuda_check_implementation at /pytorch/c10/cuda/CUDAException.cpp:42 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits, std::allocator >) + 0x80 (0x7f6ab4f7eeb0 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libc10.so)
frame #1: + 0x111c7 (0x7f6ab53a51c7 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libc10_cuda.so)
frame #2: c10d::ProcessGroupNCCL::WorkNCCL::finishedGPUExecutionInternal() const + 0x50 (0x7f6a58030640 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #3: c10d::ProcessGroupNCCL::WorkNCCL::isCompleted() + 0x68 (0x7f6a5803fe28 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #4: c10d::ProcessGroupNCCL::Watchdog::runLoop() + 0x978 (0x7f6a58042f48 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #5: c10d::ProcessGroupNCCL::Watchdog::run() + 0xd2 (0x7f6a58044ec2 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #6: + 0xecdb4 (0x7f6a3b4ecdb4 in /lib/x86_64-linux-gnu/libstdc++.so.6)
frame #7: + 0x9caa4 (0x7f6ab5e9caa4 in /lib/x86_64-linux-gnu/libc.so.6)
frame #8: + 0x129c3c (0x7f6ab5f29c3c in /lib/x86_64-linux-gnu/libc.so.6)
Exception raised from run at /pytorch/torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:2074 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits, std::allocator >) + 0x80 (0x7f6ab4f7eeb0 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libc10.so)
frame #1: + 0xe1c1a1 (0x7f6a5801c1a1 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #2: + 0x9468e6 (0x7f6a57b468e6 in /home/zjy/code/vllm-src/.venv/lib/python3.12/site-packages/torch/lib/libtorch_cuda.so)
frame #3: + 0xecdb4 (0x7f6a3b4ecdb4 in /lib/x86_64-linux-gnu/libstdc++.so.6)
frame #4: + 0x9caa4 (0x7f6ab5e9caa4 in /lib/x86_64-linux-gnu/libc.so.6)
frame #5: + 0x129c3c (0x7f6ab5f29c3c in /lib/x86_64-linux-gnu/libc.so.6)
### Before submitting a new issue...
- [x] Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the [documentation page](https://docs.vllm.ai/en/latest/), which can answer lots of frequently asked questions.