Collecting environment information...
==============================
        System Info
==============================
OS                           : Ubuntu 22.04.4 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version                : Could not collect
CMake version                : version 4.0.0
Libc version                 : glibc-2.35

==============================
       PyTorch Info
==============================
PyTorch version              : 2.11.0+cu129
Is debug build               : False
CUDA used to build PyTorch   : 12.9
ROCM used to build PyTorch   : N/A
XPU used to build PyTorch    : N/A

==============================
      Python Environment
==============================
Python version               : 3.14.4 | packaged by Anaconda, Inc. | (main, Apr 14 2026, 17:07:44) [GCC 14.3.0] (64-bit runtime)
Python platform              : Linux-4.18.0-513.5.1.el8_9.x86_64-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 12.9.86
CUDA_MODULE_LOADING set to   :
GPU models and configuration :
GPU 0: NVIDIA H100 80GB HBM3
GPU 1: NVIDIA H100 80GB HBM3
GPU 2: NVIDIA H100 80GB HBM3
GPU 3: NVIDIA H100 80GB HBM3
GPU 4: NVIDIA H100 80GB HBM3
GPU 5: NVIDIA H100 80GB HBM3
GPU 6: NVIDIA H100 80GB HBM3
GPU 7: NVIDIA H100 80GB HBM3

Nvidia driver version        : 535.161.08
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):                             112
On-line CPU(s) list:                0-111
Vendor ID:                          GenuineIntel
Model name:                         Intel(R) Xeon(R) Platinum 8480C
CPU family:                         6
Model:                              143
Thread(s) per core:                 1
Core(s) per socket:                 56
Socket(s):                          2
Stepping:                           8
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 invpcid_single intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi 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 avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req 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 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
NUMA node1 CPU(s):                  56-111
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 Retbleed:             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
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.11.post2
[pip3] numpy==2.4.6
[pip3] nvidia-cublas-cu12==12.9.1.4
[pip3] nvidia-cuda-cccl-cu12==12.9.27
[pip3] nvidia-cuda-cupti-cu12==12.9.79
[pip3] nvidia-cuda-nvcc-cu12==12.9.86
[pip3] nvidia-cuda-nvrtc-cu12==12.9.86
[pip3] nvidia-cuda-runtime-cu12==12.9.79
[pip3] nvidia-cudnn-cu12==9.17.1.4
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft-cu12==11.4.1.4
[pip3] nvidia-cufile-cu12==1.14.1.1
[pip3] nvidia-curand-cu12==10.3.10.19
[pip3] nvidia-cusolver-cu12==11.7.5.82
[pip3] nvidia-cusparse-cu12==12.5.10.65
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-cutlass-dsl==4.5.2
[pip3] nvidia-cutlass-dsl-libs-base==4.5.2
[pip3] nvidia-ml-py==13.610.43
[pip3] nvidia-nccl-cu12==2.28.9
[pip3] nvidia-nvjitlink-cu12==12.9.86
[pip3] nvidia-nvshmem-cu12==3.4.5
[pip3] nvidia-nvtx-cu12==12.9.79
[pip3] pyzmq==27.1.0
[pip3] tokenspeed-triton==3.7.10.post20260531
[pip3] torch==2.11.0+cu129
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0+cu129
[pip3] torchvision==0.26.0+cu129
[pip3] transformers==5.10.2
[pip3] triton==3.6.0
[conda] flashinfer-python         0.6.11.post2             pypi_0    pypi
[conda] numpy                     2.4.6                    pypi_0    pypi
[conda] nvidia-cublas-cu12        12.9.1.4                 pypi_0    pypi
[conda] nvidia-cuda-cccl-cu12     12.9.27                  pypi_0    pypi
[conda] nvidia-cuda-cupti-cu12    12.9.79                  pypi_0    pypi
[conda] nvidia-cuda-nvcc-cu12     12.9.86                  pypi_0    pypi
[conda] nvidia-cuda-nvrtc-cu12    12.9.86                  pypi_0    pypi
[conda] nvidia-cuda-runtime-cu12  12.9.79                  pypi_0    pypi
[conda] nvidia-cudnn-cu12         9.17.1.4                 pypi_0    pypi
[conda] nvidia-cudnn-frontend     1.18.0                   pypi_0    pypi
[conda] nvidia-cufft-cu12         11.4.1.4                 pypi_0    pypi
[conda] nvidia-cufile-cu12        1.14.1.1                 pypi_0    pypi
[conda] nvidia-curand-cu12        10.3.10.19               pypi_0    pypi
[conda] nvidia-cusolver-cu12      11.7.5.82                pypi_0    pypi
[conda] nvidia-cusparse-cu12      12.5.10.65               pypi_0    pypi
[conda] nvidia-cusparselt-cu12    0.7.1                    pypi_0    pypi
[conda] nvidia-cutlass-dsl        4.5.2                    pypi_0    pypi
[conda] nvidia-cutlass-dsl-libs-base 4.5.2                    pypi_0    pypi
[conda] nvidia-ml-py              13.610.43                pypi_0    pypi
[conda] nvidia-nccl-cu12          2.28.9                   pypi_0    pypi
[conda] nvidia-nvjitlink-cu12     12.9.86                  pypi_0    pypi
[conda] nvidia-nvshmem-cu12       3.4.5                    pypi_0    pypi
[conda] nvidia-nvtx-cu12          12.9.79                  pypi_0    pypi
[conda] pyzmq                     27.1.0                   pypi_0    pypi
[conda] tokenspeed-triton         3.7.10.post20260531          pypi_0    pypi
[conda] torch                     2.11.0+cu129             pypi_0    pypi
[conda] torch-c-dlpack-ext        0.1.5                    pypi_0    pypi
[conda] torchaudio                2.11.0+cu129             pypi_0    pypi
[conda] torchvision               0.26.0+cu129             pypi_0    pypi
[conda] transformers              5.10.2                   pypi_0    pypi
[conda] triton                    3.6.0                    pypi_0    pypi

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.22.0
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled; XPU: Disabled
GPU Topology:
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    NIC8    NIC9    NIC10   NIC11   CPU AffinityNUMA Affinity    GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    PXB     NODE    NODE    NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS     0-55    0   N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    NODE    NODE    NODE    PXB     NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS     0-55    0   N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    NODE    NODE    NODE    NODE    PXB     NODE    SYS     SYS     SYS     SYS     SYS     SYS     0-55    0   N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    NODE    NODE    NODE    NODE    NODE    PXB     SYS     SYS     SYS     SYS     SYS     SYS     0-55    0   N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    SYS     SYS     SYS     SYS     SYS     SYS     PXB     NODE    NODE    NODE    NODE    NODE    56-111  1   N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    SYS     SYS     SYS     SYS     SYS     SYS     NODE    NODE    NODE    PXB     NODE    NODE    56-111  1   N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    SYS     SYS     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PXB     NODE    56-111  1   N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      SYS     SYS     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    NODE    PXB     56-111  1   N/A
NIC0    PXB     NODE    NODE    NODE    SYS     SYS     SYS     SYS      X      NODE    NODE    NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS
NIC1    NODE    NODE    NODE    NODE    SYS     SYS     SYS     SYS     NODE     X      PIX     NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS
NIC2    NODE    NODE    NODE    NODE    SYS     SYS     SYS     SYS     NODE    PIX      X      NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS
NIC3    NODE    PXB     NODE    NODE    SYS     SYS     SYS     SYS     NODE    NODE    NODE     X      NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS
NIC4    NODE    NODE    PXB     NODE    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE     X      NODE    SYS     SYS     SYS     SYS     SYS     SYS
NIC5    NODE    NODE    NODE    PXB     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    NODE     X      SYS     SYS     SYS     SYS     SYS     SYS
NIC6    SYS     SYS     SYS     SYS     PXB     NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS      X      NODE    NODE    NODE    NODE    NODE
NIC7    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS     NODE     X      PIX     NODE    NODE    NODE
NIC8    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS     NODE    PIX      X      NODE    NODE    NODE
NIC9    SYS     SYS     SYS     SYS     NODE    PXB     NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS     NODE    NODE    NODE     X      NODE    NODE
NIC10   SYS     SYS     SYS     SYS     NODE    NODE    PXB     NODE    SYS     SYS     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE     X      NODE
NIC11   SYS     SYS     SYS     SYS     NODE    NODE    NODE    PXB     SYS     SYS     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    NODE     X

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

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1
  NIC2: mlx5_2
  NIC3: mlx5_3
  NIC4: mlx5_4
  NIC5: mlx5_5
  NIC6: mlx5_6
  NIC7: mlx5_7
  NIC8: mlx5_8
  NIC9: mlx5_9
  NIC10: mlx5_10
  NIC11: mlx5_11

==============================
     Environment Variables
==============================
VLLM_CACHE_ROOT=/work/HHRI-AI/vllm_cache
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_root

---

if __name__ == "__main__":
    from vllm import LLM, SamplingParams
    from transformers import AutoTokenizer

    # Initialize Qwen3-8B
    llm = LLM(
        model="Qwen/Qwen3.5-9B",
        reasoning_parser="qwen3",   # Specific parser for Qwen3's thought tokens
        max_model_len=32768,         # Qwen3 supports up to 131K, adjust as needed
        tensor_parallel_size=2,
        gpu_memory_utilization=0.9,
        download_dir="/xxx/.cache",
        # add flags
        trust_remote_code=True,
        max_num_seqs=64
    )
    tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-9B")
    sampling_params = SamplingParams(
        temperature=0.0, 
        max_tokens=8192, # 29000
    )

---

(jf_vllm) root:tw_eval# CUDA_VISIBLE_DEVICES=3,4 python cv_eval.py
INFO 06-09 01:30:38 [utils.py:278] non-default args: {'trust_remote_code': True, 'download_dir': '/work/HHRI-AI/joseph/.cache', 'max_model_len': 32768, 'tensor_parallel_size': 2, 'gpu_memory_utilization': 0.9, 'max_num_seqs': 64, 'disable_log_stats': True, 'reasoning_parser': 'qwen3', 'model': 'Qwen/Qwen3.5-9B'}
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
INFO 06-09 01:30:40 [model.py:617] Resolved architecture: Qwen3_5ForConditionalGeneration
INFO 06-09 01:30:40 [model.py:1752] Using max model len 32768
INFO 06-09 01:30:40 [scheduler.py:239] Chunked prefill is enabled with max_num_batched_tokens=16384.
INFO 06-09 01:30:40 [vllm.py:977] Asynchronous scheduling is enabled.
INFO 06-09 01:30:40 [kernel.py:270] Final IR op priority after setting platform defaults: IrOpPriorityConfig(rms_norm=['native'], fused_add_rms_norm=['native'])
<frozen importlib._bootstrap_external>:1061: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1061: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
INFO 06-09 01:30:46 [compilation.py:312] Enabled custom fusions: allreduce_rms
[transformers] `Qwen2VLImageProcessorFast` is deprecated. The `Fast` suffix for image processors has been removed; use `Qwen2VLImageProcessor` instead.
[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
WARNING 06-09 01:31:05 [system_utils.py:157] We must use the `spawn` multiprocessing start method. Overriding VLLM_WORKER_MULTIPROC_METHOD to 'spawn'. See https://docs.vllm.ai/en/latest/usage/troubleshooting.html#python-multiprocessing for more information. Reasons: CUDA is initialized
(EngineCore pid=2951629) INFO 06-09 01:31:51 [core.py:112] Initializing a V1 LLM engine (v0.22.0) with config: model='Qwen/Qwen3.5-9B', speculative_config=None, tokenizer='Qwen/Qwen3.5-9B', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, tokenizer_revision=None, trust_remote_code=True, dtype=torch.bfloat16, max_seq_len=32768, download_dir='/work/HHRI-AI/joseph/.cache', load_format=auto, tensor_parallel_size=2, pipeline_parallel_size=1, data_parallel_size=1, decode_context_parallel_size=1, dcp_comm_backend=ag_rs, disable_custom_all_reduce=False, quantization=None, quantization_config=None, enforce_eager=False, enable_return_routed_experts=False, kv_cache_dtype=auto, device_config=cuda, structured_outputs_config=StructuredOutputsConfig(backend='auto', disable_any_whitespace=False, disable_additional_properties=False, reasoning_parser='qwen3', reasoning_parser_plugin='', enable_in_reasoning=False), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, kv_cache_metrics=False, kv_cache_metrics_sample=0.01, cudagraph_metrics=False, enable_layerwise_nvtx_tracing=False, enable_mfu_metrics=False, enable_mm_processor_stats=False, enable_logging_iteration_details=False), seed=0, served_model_name=Qwen/Qwen3.5-9B, enable_prefix_caching=False, enable_chunked_prefill=True, pooler_config=None, compilation_config={'mode': <CompilationMode.VLLM_COMPILE: 3>, 'debug_dump_path': None, 'cache_dir': '', 'compile_cache_save_format': 'binary', 'backend': 'inductor', 'custom_ops': ['none'], 'ir_enable_torch_wrap': True, 'splitting_ops': ['vllm::unified_attention_with_output', 'vllm::unified_mla_attention_with_output', 'vllm::mamba_mixer2', 'vllm::mamba_mixer', 'vllm::short_conv', 'vllm::linear_attention', 'vllm::plamo2_mamba_mixer', 'vllm::qwen_gdn_attention_core', 'vllm::gdn_attention_core_xpu', 'vllm::olmo_hybrid_gdn_full_forward', 'vllm::kda_attention', 'vllm::sparse_attn_indexer', 'vllm::rocm_aiter_sparse_attn_indexer', 'vllm::deepseek_v4_attention', 'vllm::unified_kv_cache_update', 'vllm::unified_mla_kv_cache_update'], 'compile_mm_encoder': False, 'cudagraph_mm_encoder': False, 'encoder_cudagraph_token_budgets': [], 'encoder_cudagraph_max_vision_items_per_batch': 0, 'encoder_cudagraph_max_frames_per_batch': None, 'compile_sizes': [], 'compile_ranges_endpoints': [8192, 16384], 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'size_asserts': False, 'alignment_asserts': False, 'scalar_asserts': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': <CUDAGraphMode.FULL_AND_PIECEWISE: (2, 1)>, 'cudagraph_num_of_warmups': 1, 'cudagraph_capture_sizes': [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128], 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': False, 'pass_config': {'fuse_norm_quant': False, 'fuse_act_quant': False, 'fuse_attn_quant': False, 'enable_sp': False, 'fuse_gemm_comms': False, 'fuse_allreduce_rms': True, 'fuse_rope_kvcache_cat_mla': False, 'fuse_act_padding': False}, 'max_cudagraph_capture_size': 128, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>, 'evaluate_guards': False, 'assume_32_bit_indexing': False}, 'local_cache_dir': None, 'fast_moe_cold_start': False, 'static_all_moe_layers': []}, kernel_config=KernelConfig(ir_op_priority=IrOpPriorityConfig(rms_norm=['native'], fused_add_rms_norm=['native']), enable_flashinfer_autotune=True, moe_backend='auto', linear_backend='auto')
(EngineCore pid=2951629) WARNING 06-09 01:31:51 [multiproc_executor.py:1029] Reducing Torch parallelism from 108 threads to 1 to avoid unnecessary CPU contention. Set OMP_NUM_THREADS in the external environment to tune this value as needed.
(EngineCore pid=2951629) INFO 06-09 01:31:51 [multiproc_executor.py:139] DP group leader: node_rank=0, node_rank_within_dp=0, master_addr=127.0.0.1, mq_connect_ip=192.168.23.95 (local), world_size=2, local_world_size=2
[transformers] `Qwen2VLImageProcessorFast` is deprecated. The `Fast` suffix for image processors has been removed; use `Qwen2VLImageProcessor` instead.
[transformers] `Qwen2VLImageProcessorFast` is deprecated. The `Fast` suffix for image processors has been removed; use `Qwen2VLImageProcessor` instead.
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
(Worker pid=2952888) INFO 06-09 01:32:44 [parallel_state.py:1422] world_size=2 rank=0 local_rank=0 distributed_init_method=tcp://127.0.0.1:44339 backend=nccl
(Worker pid=2952889) INFO 06-09 01:32:44 [parallel_state.py:1422] world_size=2 rank=1 local_rank=1 distributed_init_method=tcp://127.0.0.1:44339 backend=nccl
(Worker pid=2952889) <frozen importlib._bootstrap_external>:1061: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
(Worker pid=2952888) <frozen importlib._bootstrap_external>:1061: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
(Worker pid=2952889) <frozen importlib._bootstrap_external>:1061: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
(Worker pid=2952888) <frozen importlib._bootstrap_external>:1061: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
(Worker pid=2952888) INFO 06-09 01:32:47 [pynccl.py:113] vLLM is using nccl==2.28.9
(Worker pid=2952888) INFO 06-09 01:32:50 [cuda_communicator.py:232] Using ['CUSTOM', 'SYMM_MEM', 'PYNCCL'] all-reduce backends (in dispatch order) for group 'tp:0' out of potential backends: ['NCCL_SYMM_MEM', 'QUICK_REDUCE', 'FLASHINFER', 'CUSTOM', 'SYMM_MEM', 'PYNCCL'].
(Worker pid=2952888) INFO 06-09 01:32:50 [parallel_state.py:1735] rank 0 in world size 2 is assigned as DP rank 0, PP rank 0, PCP rank 0, TP rank 0, EP rank N/A, EPLB rank N/A
(Worker pid=2952888) INFO 06-09 01:32:50 [topk_topp_sampler.py:45] Using FlashInfer for top-p & top-k sampling.
(Worker pid=2952888) [transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
(Worker pid=2952889) [transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:05 [gpu_model_runner.py:5037] Starting to load model Qwen/Qwen3.5-9B...
(Worker_TP0 pid=2952888) INFO 06-09 01:33:06 [cuda.py:433] Using backend AttentionBackendEnum.FLASH_ATTN for vit attention
(Worker_TP0 pid=2952888) INFO 06-09 01:33:06 [mm_encoder_attention.py:372] Using AttentionBackendEnum.FLASH_ATTN for MMEncoderAttention.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:06 [qwen_gdn_linear_attn.py:228] Using FlashInfer GDN prefill kernel (requested=auto, head_k_dim=None).
(Worker_TP0 pid=2952888) WARNING 06-09 01:33:06 [qwen_gdn_linear_attn.py:235] FlashInfer GDN prefill is JIT-compiled; first run may take a while. Set --gdn-prefill-backend triton to skip JIT.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:06 [cuda.py:378] Using FLASH_ATTN attention backend out of potential backends: ['FLASH_ATTN', 'FLASHINFER', 'TRITON_ATTN', 'FLEX_ATTENTION'].
(Worker_TP0 pid=2952888) INFO 06-09 01:33:06 [flash_attn.py:636] Using FlashAttention version 3
(Worker_TP0 pid=2952888) INFO 06-09 01:33:07 [weight_utils.py:922] Filesystem type for checkpoints: WEKAFS. Checkpoint size: 17.98 GiB. Available RAM: 1830.66 GiB.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:07 [weight_utils.py:945] Auto-prefetch is disabled because the filesystem (WEKAFS) is not a recognized network FS (NFS/Lustre). If you want to force prefetching, start vLLM with --safetensors-load-strategy=prefetch.
Loading safetensors checkpoint shards:   0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards:  25% Completed | 1/4 [00:01<00:03,  1.04s/it]
Loading safetensors checkpoint shards:  50% Completed | 2/4 [00:02<00:02,  1.21s/it]
Loading safetensors checkpoint shards:  75% Completed | 3/4 [00:03<00:01,  1.28s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:04<00:00,  1.14s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:04<00:00,  1.16s/it]
(Worker_TP0 pid=2952888)
(Worker_TP0 pid=2952888) INFO 06-09 01:33:12 [default_loader.py:397] Loading weights took 4.77 seconds
(Worker_TP1 pid=2952889) INFO 06-09 01:33:13 [interface.py:649] Setting attention block size to 528 tokens to ensure that attention page size is >= mamba page size.
(Worker_TP1 pid=2952889) INFO 06-09 01:33:13 [interface.py:673] Padding mamba page size by 0.76% to ensure that mamba page size and attention page size are exactly equal.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:13 [gpu_model_runner.py:5132] Model loading took 8.91 GiB memory and 6.532624 seconds
(Worker_TP0 pid=2952888) INFO 06-09 01:33:13 [interface.py:649] Setting attention block size to 528 tokens to ensure that attention page size is >= mamba page size.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:13 [interface.py:673] Padding mamba page size by 0.76% to ensure that mamba page size and attention page size are exactly equal.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:14 [gpu_model_runner.py:6136] Encoder cache will be initialized with a budget of 16384 tokens, and profiled with 1 image items of the maximum feature size.
(Worker_TP0 pid=2952888) INFO 06-09 01:33:21 [backends.py:1089] Using cache directory: /work/HHRI-AI/vllm_cache/torch_compile_cache/8c560267da/rank_0_0/backbone for vLLM's torch.compile
(Worker_TP0 pid=2952888) INFO 06-09 01:33:21 [backends.py:1148] Dynamo bytecode transform time: 5.00 s
(Worker_TP0 pid=2952888) INFO 06-09 01:33:21 [flashinfer_all_reduce.py:111] Auto-selected flashinfer allreduce backend: trtllm
(EngineCore pid=2951629) INFO 06-09 01:34:14 [shm_broadcast.py:698] No available shared memory broadcast block found in 60 seconds. This typically happens when some processes are hanging or doing some time-consuming work (e.g. compilation, weight/kv cache quantization).
(EngineCore pid=2951629) INFO 06-09 01:35:14 [shm_broadcast.py:698] No available shared memory broadcast block found in 60 seconds. This typically happens when some processes are hanging or doing some time-consuming work (e.g. compilation, weight/kv cache quantization).
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] Failed to initialize FlashInfer All Reduce workspace: Ninja build failed. Ninja output:
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] ninja: Entering directory `/work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm'
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] [1/1] c++ /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/csrc_trtllm_allreduce.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/csrc_trtllm_allreduce_fusion.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/csrc_trtllm_moe_allreduce_fusion.cuda.o -shared -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64 -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64/stubs -lcudart -lcuda -o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/trtllm_comm.so
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] FAILED: [code=1] /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/trtllm_comm.so
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] c++ /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/csrc_trtllm_allreduce.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/csrc_trtllm_allreduce_fusion.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/csrc_trtllm_moe_allreduce_fusion.cuda.o -shared -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64 -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64/stubs -lcudart -lcuda -o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/trtllm_comm/trtllm_comm.so
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] /usr/bin/ld: cannot find -lcudart: No such file or directory
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] collect2: error: ld returned 1 exit status
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] ninja: build stopped: subcommand failed.
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:74] .
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [flashinfer_all_reduce.py:156] Failed to initialize FlashInfer Allreduce norm fusion workspace with backend=trtllm
(Worker_TP0 pid=2952888) WARNING 06-09 01:36:04 [allreduce_rms_fusion.py:820] Failed to initialize Flashinfer allreduce workspace. Flashinfer allreduce-norm fusion will be disabled.
(Worker_TP1 pid=2952889) INFO 06-09 01:36:10 [decorators.py:311] Directly load AOT compilation from path /work/HHRI-AI/vllm_cache/torch_compile_cache/torch_aot_compile/669a41c51d6f56926c5bfaf87c3dfa8825b546e0c29f16719093bc42a07f848e/rank_1_0/model
(Worker_TP0 pid=2952888) INFO 06-09 01:36:11 [backends.py:292] Directly load the compiled graph(s) for compile range (1, 8192) from the cache, took 6.333 s
(Worker_TP0 pid=2952888) INFO 06-09 01:36:11 [backends.py:292] Directly load the compiled graph(s) for compile range (8193, 16384) from the cache, took 6.432 s
(Worker_TP0 pid=2952888) INFO 06-09 01:36:11 [decorators.py:311] Directly load AOT compilation from path /work/HHRI-AI/vllm_cache/torch_compile_cache/torch_aot_compile/669a41c51d6f56926c5bfaf87c3dfa8825b546e0c29f16719093bc42a07f848e/rank_0_0/model
(Worker_TP0 pid=2952888) INFO 06-09 01:36:11 [monitor.py:53] torch.compile took 174.87 s in total
(EngineCore pid=2951629) INFO 06-09 01:36:14 [shm_broadcast.py:698] No available shared memory broadcast block found in 60 seconds. This typically happens when some processes are hanging or doing some time-consuming work (e.g. compilation, weight/kv cache quantization).
(EngineCore pid=2951629) INFO 06-09 01:37:14 [shm_broadcast.py:698] No available shared memory broadcast block found in 60 seconds. This typically happens when some processes are hanging or doing some time-consuming work (e.g. compilation, weight/kv cache quantization).

---

(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return self._call_impl(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return forward_call(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/sample/ops/topk_topp_sampler.py", line 157, in forward_cuda
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return flashinfer_sample(logits.contiguous(), k, p, generators), None
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/sample/ops/topk_topp_sampler.py", line 457, in flashinfer_sample
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     next_token_ids = flashinfer.sampling.top_k_top_p_sampling_from_logits(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         logits, k, p, deterministic=True
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     )
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/api_logging.py", line 1546, in _auto_dump_wrapper
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return _inner(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/sampling.py", line 1414, in top_k_top_p_sampling_from_logits
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     masked_logits = top_k_mask_logits(logits, top_k)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/api_logging.py", line 1546, in _auto_dump_wrapper
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return _inner(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/sampling.py", line 1820, in top_k_mask_logits
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return get_sampling_module().top_k_mask_logits(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~~~~~^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/sampling.py", line 68, in get_sampling_module
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     module = gen_sampling_module().build_and_load()
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/core.py", line 316, in build_and_load
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     self.build(verbose, need_lock=False)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/core.py", line 302, in build
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     run_ninja(self.build_dir, self.ninja_path, verbose)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/cpp_ext.py", line 380, in run_ninja
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     raise RuntimeError(msg) from e
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] RuntimeError: Ninja build failed. Ninja output:
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] ninja: Entering directory `/work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling'
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] [1/1] c++ /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_sampling.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_renorm.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_flashinfer_sampling_binding.cuda.o -shared -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64 -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64/stubs -lcudart -lcuda -o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/sampling.so
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] FAILED: [code=1] /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/sampling.so
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] c++ /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_sampling.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_renorm.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_flashinfer_sampling_binding.cuda.o -shared -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64 -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64/stubs -lcudart -lcuda -o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/sampling.so
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] /usr/bin/ld: cannot find -lcudart: No such file or directory
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] collect2: error: ld returned 1 exit status
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] ninja: build stopped: subcommand failed.
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] Traceback (most recent call last):
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/cpp_ext.py", line 368, in run_ninja
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     subprocess.run(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ~~~~~~~~~~~~~~^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         command,
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         ^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ...<4 lines>...
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         text=True,
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         ^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     )
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/subprocess.py", line 578, in run
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     raise CalledProcessError(retcode, process.args,
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]                              output=stdout, stderr=stderr)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] subprocess.CalledProcessError: Command '['ninja', '-v', '-C', '/work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling', '-f', '/work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/build.ninja']' returned non-zero exit status 1.
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] The above exception was the direct cause of the following exception:
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] Traceback (most recent call last):
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/executor/multiproc_executor.py", line 957, in worker_busy_loop
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     output = func(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/utils/_contextlib.py", line 124, in decorate_context
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return func(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/worker/gpu_worker.py", line 396, in determine_available_memory
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     self.model_runner.profile_run()
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/worker/gpu_model_runner.py", line 6171, in profile_run
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     output = self._dummy_sampler_run(last_hidden_states)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/utils/_contextlib.py", line 124, in decorate_context
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return func(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/worker/gpu_model_runner.py", line 5983, in _dummy_sampler_run
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     raise e
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/worker/gpu_model_runner.py", line 5951, in _dummy_sampler_run
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     sampler_output = self.sampler(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         logits=logits, sampling_metadata=dummy_metadata
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     )
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return self._call_impl(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return forward_call(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/sample/sampler.py", line 97, in forward
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     sampled, processed_logprobs = self.sample(logits, sampling_metadata)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]                                   ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/sample/sampler.py", line 281, in sample
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     random_sampled, processed_logprobs = self.topk_topp_sampler(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]                                          ~~~~~~~~~~~~~~~~~~~~~~^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         logits,
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         ^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ...<2 lines>...
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         sampling_metadata.top_p,
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         ^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     )
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return self._call_impl(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return forward_call(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/sample/ops/topk_topp_sampler.py", line 157, in forward_cuda
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return flashinfer_sample(logits.contiguous(), k, p, generators), None
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/sample/ops/topk_topp_sampler.py", line 457, in flashinfer_sample
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     next_token_ids = flashinfer.sampling.top_k_top_p_sampling_from_logits(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]         logits, k, p, deterministic=True
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     )
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/api_logging.py", line 1546, in _auto_dump_wrapper
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return _inner(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/sampling.py", line 1414, in top_k_top_p_sampling_from_logits
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     masked_logits = top_k_mask_logits(logits, top_k)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/api_logging.py", line 1546, in _auto_dump_wrapper
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return _inner(*args, **kwargs)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/sampling.py", line 1820, in top_k_mask_logits
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     return get_sampling_module().top_k_mask_logits(
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]            ~~~~~~~~~~~~~~~~~~~^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/sampling.py", line 68, in get_sampling_module
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     module = gen_sampling_module().build_and_load()
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/core.py", line 316, in build_and_load
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     self.build(verbose, need_lock=False)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/core.py", line 302, in build
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     run_ninja(self.build_dir, self.ninja_path, verbose)
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]   File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/flashinfer/jit/cpp_ext.py", line 380, in run_ninja
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]     raise RuntimeError(msg) from e
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] RuntimeError: Ninja build failed. Ninja output:
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] ninja: Entering directory `/work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling'
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] [1/1] c++ /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_sampling.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_renorm.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_flashinfer_sampling_binding.cuda.o -shared -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64 -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64/stubs -lcudart -lcuda -o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/sampling.so
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] FAILED: [code=1] /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/sampling.so
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] c++ /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_sampling.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_renorm.cuda.o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/csrc_flashinfer_sampling_binding.cuda.o -shared -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64 -L/work/HHRI-AI/anaconda/envs/vllm_gemma4/lib64/stubs -lcudart -lcuda -o /work/HHRI-AI/root_cache/flashinfer/0.6.11.post2/90a/cached_ops/sampling/sampling.so
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] /usr/bin/ld: cannot find -lcudart: No such file or directory
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] collect2: error: ld returned 1 exit status
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962] ninja: build stopped: subcommand failed.
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]
(Worker_TP1 pid=2911594) ERROR 06-09 01:15:22 [multiproc_executor.py:962]
(Worker_TP1 pid=2911594) INFO 06-09 01:15:22 [multiproc_executor.py:872] WorkerProc shutting down.
Traceback (most recent call last):
  File "/work/HHRI-AI/joseph/tw_eval/cv_eval.py", line 135, in <module>
    llm = LLM(
        #model="/work/HHRI-AI/Foxbrain/rick/FoxBrain_Group/Pretrained_weights/VLM-model/Alibaba/qwen35/Qwen3.6-35B-A3B", #"Qwen/Qwen3-8B"
    ...<10 lines>...
        #enable_expert_parallel=True
    )
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/entrypoints/llm.py", line 349, in __init__
    self.llm_engine = LLMEngine.from_engine_args(
                      ~~~~~~~~~~~~~~~~~~~~~~~~~~^
        engine_args=engine_args, usage_context=UsageContext.LLM_CLASS
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/llm_engine.py", line 170, in from_engine_args
    return cls(
        vllm_config=vllm_config,
    ...<4 lines>...
        multiprocess_mode=enable_multiprocessing,
    )
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/llm_engine.py", line 104, in __init__
    self.engine_core = EngineCoreClient.make_client(
                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
        multiprocess_mode=multiprocess_mode,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<3 lines>...
        log_stats=self.log_stats,
        ^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/core_client.py", line 102, in make_client
    return SyncMPClient(vllm_config, executor_class, log_stats)
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
    return func(*args, **kwargs)
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/core_client.py", line 755, in __init__
    super().__init__(
    ~~~~~~~~~~~~~~~~^
        asyncio_mode=False,
        ^^^^^^^^^^^^^^^^^^^
    ...<2 lines>...
        log_stats=log_stats,
        ^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/core_client.py", line 567, in __init__
    with launch_core_engines(
         ~~~~~~~~~~~~~~~~~~~^
        vllm_config, executor_class, log_stats, addresses
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ) as (engine_manager, coordinator, addresses, tensor_queue):
    ^
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/contextlib.py", line 148, in __exit__
    next(self.gen)
    ~~~~^^^^^^^^^^
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/utils.py", line 1150, in launch_core_engines
    wait_for_engine_startup(
    ~~~~~~~~~~~~~~~~~~~~~~~^
        handshake_socket,
        ^^^^^^^^^^^^^^^^^
    ...<6 lines>...
        coordinator.proc if coordinator else None,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/work/HHRI-AI/anaconda/envs/jf_vllm/lib/python3.14/site-packages/vllm/v1/engine/utils.py", line 1209, in wait_for_engine_startup
    raise RuntimeError(
    ...<3 lines>...
    )
RuntimeError: Engine core initialization failed. See root cause above. Failed core proc(s): {}