[Bug]: DeepSeek V3.2 running with MTP will cause DeepseekV32IndexerMetadata parsing error

[RekingQin]

Your current environment

The output of python collect_env.py

Collecting environment information...
==============================
        System Info
==============================
OS                           : Ubuntu 22.04.5 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version                : Could not collect
CMake version                : version 4.1.0
Libc version                 : glibc-2.35

==============================
       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               : 3.12.11 (main, Jun  4 2025, 08:56:18) [GCC 11.4.0] (64-bit runtime)
Python platform              : Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : Could not collect
CUDA_MODULE_LOADING set to   : LAZY
GPU models and configuration : 
GPU 0: NVIDIA H20
GPU 1: NVIDIA H20
GPU 2: NVIDIA H20
GPU 3: NVIDIA H20
GPU 4: NVIDIA H20
GPU 5: NVIDIA H20
GPU 6: NVIDIA H20
GPU 7: NVIDIA H20

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):                          384
On-line CPU(s) list:             0-383
Vendor ID:                       AuthenticAMD
BIOS Vendor ID:                  Advanced Micro Devices, Inc.
Model name:                      AMD EPYC 9K84 96-Core Processor
BIOS Model name:                 AMD EPYC 9K84 96-Core Processor                
CPU family:                      25
Model:                           17
Thread(s) per core:              2
Core(s) per socket:              96
Socket(s):                       2
Stepping:                        1
Frequency boost:                 enabled
CPU max MHz:                     2600.0000
CPU min MHz:                     1500.0000
BogoMIPS:                        5199.96
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 invpcid_single hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local avx512_bf16 clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d
Virtualization:                  AMD-V
L1d cache:                       6 MiB (192 instances)
L1i cache:                       6 MiB (192 instances)
L2 cache:                        192 MiB (192 instances)
L3 cache:                        768 MiB (24 instances)
NUMA node(s):                    2
NUMA node0 CPU(s):               0-95,192-287
NUMA node1 CPU(s):               96-191,288-383
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 and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Mitigation; Retpolines, IBPB conditional, IBRS_FW, STIBP always-on, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.4.0
[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==1.15.0
[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-cutlass-dsl==4.2.1
[pip3] nvidia-ml-py==13.580.82
[pip3] nvidia-nccl-cu12==2.27.3
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pyzmq==27.1.0
[pip3] torch==2.8.0+cu128
[pip3] torchaudio==2.8.0+cu128
[pip3] torchvision==0.23.0+cu128
[pip3] transformers==4.57.0
[pip3] triton==3.4.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.11.1rc1.dev417+gac8f57d00 (git sha: ac8f57d00)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled
GPU Topology:
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    NIC8    NIC9    NIC10   NIC11   NIC12   NIC13   NIC14   NIC15   NIC16   NIC17   NIC18   NIC19   NIC20   NIC21   NIC22   NIC23   NIC24   NIC25   NIC26   NIC27   NIC28   NIC29   NIC30   NIC31    NIC32   NIC33   NIC34   NIC35   NIC36   NIC37   NIC38   NIC39   NIC40   NIC41   CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     PIX     NODE    NODE    NODE    SYS     SYS     SYS     SYS     0-95,192-287    0               N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     NODE    NODE    PHB     PIX     SYS     SYS     SYS     SYS     0-95,192-287    0               N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     NODE    NODE    PIX     PHB     SYS     SYS     SYS     SYS     0-95,192-287    0               N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     NODE    PIX     NODE    NODE    SYS     SYS     SYS     SYS     0-95,192-287    0               N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     NODE    NODE    PIX     NODE    96-191,288-383  1               N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     NODE    PIX     NODE    NODE    96-191,288-383  1               N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     PHB     NODE    NODE    PIX     96-191,288-383  1               N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     PIX     NODE    NODE    PHB     96-191,288-383  1               N/A
NIC0    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE     X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC1    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC2    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC3    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC4    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC5    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC6    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC7    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC8    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC9    SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC10   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC11   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC12   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC13   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC14   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC15   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC16   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC17   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC18   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC19   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC20   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC21   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC22   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC23   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC24   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC25   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC26   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC27   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC28   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC29   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX     PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC30   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X      PIX      PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC31   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      X       PIX     PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC32   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX       X      PIX     SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC33   SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE    PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX     PIX      PIX      X      SYS     SYS     SYS     SYS     NODE    NODE    NODE    NODE
NIC34   PIX     NODE    NODE    NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS      X      NODE    NODE    NODE    SYS     SYS     SYS     SYS
NIC35   NODE    NODE    NODE    PIX     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     NODE     X      NODE    NODE    SYS     SYS     SYS     SYS
NIC36   NODE    PHB     PIX     NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     NODE    NODE     X      PHB     SYS     SYS     SYS     SYS
NIC37   NODE    PIX     PHB     NODE    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      SYS     SYS     NODE    NODE    PHB      X      SYS     SYS     SYS     SYS
NIC38   SYS     SYS     SYS     SYS     NODE    NODE    PHB     PIX     NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS      X      NODE    NODE    PHB
NIC39   SYS     SYS     SYS     SYS     NODE    PIX     NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     NODE     X      NODE    NODE
NIC40   SYS     SYS     SYS     SYS     PIX     NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     NODE    NODE     X      NODE
NIC41   SYS     SYS     SYS     SYS     NODE    NODE    PIX     PHB     NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE    NODE     NODE    NODE    SYS     SYS     SYS     SYS     PHB     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:         NIC41   CPU Affinity    NUMA Affinity   GPU NUMA ID
  NIC1: MA ID
  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
  NIC12: mlx5_12
  NIC13: mlx5_13
  NIC14: mlx5_14
  NIC15: mlx5_15
  NIC16: mlx5_16
  NIC17: mlx5_17
  NIC18: mlx5_18
  NIC19: mlx5_19
  NIC20: mlx5_20
  NIC21: mlx5_21
  NIC22: mlx5_22
  NIC23: mlx5_23
  NIC24: mlx5_24
  NIC25: mlx5_25
  NIC26: mlx5_26
  NIC27: mlx5_27
  NIC28: mlx5_28
  NIC29: mlx5_29
  NIC30: mlx5_30
  NIC31: mlx5_31
  NIC32: mlx5_32
  NIC33: mlx5_33
  NIC34: mlx5_bond_1
  NIC35: mlx5_bond_2
  NIC36: mlx5_bond_3
  NIC37: mlx5_bond_4
  NIC38: mlx5_bond_5
  NIC39: mlx5_bond_6
  NIC40: mlx5_bond_7
  NIC41: mlx5_bond_8

==============================
     Environment Variables
==============================
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

start the service with command

vllm serve ${MODEL_PATH} \
    --trust-remote-code \
    --served-model-name deepseek_v32  \
    --max-model-len 32000 \
    --max-num-seqs 32 \
    --gpu-memory-utilization 0.9 \
    -tp 8 \
    --max-num-batched-tokens 8000 \
    --kv-cache-dtype bfloat16 \
    --speculative_config '{"num_speculative_tokens":1,"method":"deepseek_mtp"}' \
    --no-enable-prefix-caching

Firstly, I come across the error

^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694] WorkerProc hit an exception.^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694] Traceback (most recent call last):^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/multiproc_executor.py", line 689, in worker_busy_loop^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     output = func(*args, **kwargs)^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]              ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_worker.py", line 361, in compile_or_warm_up_model^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     cuda_graph_memory_bytes = self.model_runner.capture_model()^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 3761, in capture_model^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     self._capture_cudagraphs(^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 3864, in _capture_cudagraphs^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     self._dummy_run(^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     return func(*args, **kwargs)^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 3473, in _dummy_run^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     self.drafter.dummy_run(num_tokens, use_cudagraphs=use_cudagraphs)^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     return func(*args, **kwargs)^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/spec_decode/eagle.py", line 1067, in dummy_run^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]     if use_cudagraphs and num_tokens <= self.cudagraph_batch_sizes[-1]:^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694]                                         ~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^M
^[[1;36m(Worker_TP0 pid=131406)^[[0;0m ERROR 10-13 19:43:47 [v1/executor/multiproc_executor.py:694] IndexError: list index out of range^M

then I change the code "vllm/config/speculative.py+241" to be

self.enforce_eager = False

solved the error reported

then, continue start the service will result in

^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694] WorkerProc hit an exception.^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694] Traceback (most recent call last):^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/multiproc_executor.py", line 689, in worker_busy_loop^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     output = func(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]              ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/worker_base.py", line 373, in execute_model^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self.worker.execute_model(scheduler_output, *args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return func(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_worker.py", line 476, in execute_model^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     output = self.model_runner.execute_model(scheduler_output, intermediate_tensors)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return func(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 2620, in execute_model^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     propose_draft_token_ids(sampler_output.sampled_token_ids)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 2585, in propose_draft_token_ids^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     self._draft_token_ids = self.propose_draft_token_ids(^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 2821, in propose_draft_token_ids^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     draft_token_ids = self.drafter.propose(^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]                       ^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/spec_decode/eagle.py", line 306, in propose^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     ret_hidden_states = self.model(^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]                         ^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/compilation/decorators.py", line 413, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     model_output = self.forward(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/deepseek_mtp.py", line 178, in forward^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     def forward(^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 929, in _fn^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return fn(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/compilation/caching.py", line 52, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self.optimized_call(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 848, in call_wrapped^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self._wrapped_call(self, *args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 424, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     raise e^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 411, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return super(self.cls, obj).__call__(*args, **kwargs)  # type: ignore[misc]^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self._call_impl(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1784, in _call_impl^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return forward_call(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "<eval_with_key>.186", line 45, in forward^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     submod_3 = self.submod_3(getitem_4, s72, getitem_5, getitem_6, submod_2);  getitem_4 = getitem_5 = getitem_6 = submod_3 = None^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 848, in call_wrapped^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self._wrapped_call(self, *args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 424, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     raise e^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 411, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return super(self.cls, obj).__call__(*args, **kwargs)  # type: ignore[misc]^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self._call_impl(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1784, in _call_impl^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return forward_call(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "<eval_with_key>.184", line 5, in forward^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     unified_mla_attention_with_output = torch.ops.vllm.unified_mla_attention_with_output(q, x_19, key_rot_1, output_8, 'model.layers.7.self_attn.attn');  q = x_19 = key_rot_1 = output_8 = unified_mla_attention_with_output = None^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]                                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/_ops.py", line 1243, in __call__^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     return self._op(*args, **kwargs)^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/attention/layer.py", line 991, in unified_mla_attention_with_output^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     self.impl.forward(^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/attention/backends/mla/flashmla_sparse.py", line 509, in forward^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     attn_metadata.req_id_per_token,^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694]     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^M
^[[1;36m(Worker_TP6 pid=100106)^[[0;0m ERROR 10-13 17:28:00 [v1/executor/multiproc_executor.py:694] AttributeError: 'DeepseekV32IndexerMetadata' object has no attribute 'req_id_per_token'^M

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[xiaohajiayou]

This bug is fixed in the latest vLLM (see PR #25982).
Root cause in older builds: the EAGLE drafter (MTP path) accessed cudagraph_batch_sizes[-1] when the list was empty during warmup/CUDA graph capture.
Fix: the drafter now decides CUDA graph usage internally and guards empty lists; gpu_model_runner no longer passes a use_cudagraphs flag into dummy_run.
Might can upgrade to the latest release/main? Temporary workaround on older versions: set speculative enforce_eager=true for MTP.

[RekingQin]

This bug is fixed in the latest vLLM (see PR #25982). Root cause in older builds: the EAGLE drafter (MTP path) accessed cudagraph_batch_sizes[-1] when the list was empty during warmup/CUDA graph capture. Fix: the drafter now decides CUDA graph usage internally and guards empty lists; gpu_model_runner no longer passes a use_cudagraphs flag into dummy_run. Might can upgrade to the latest release/main? Temporary workaround on older versions: set speculative enforce_eager=true for MTP.

yeah, we tested the image generated from date 25/10/08, it's really work, but for the yesterday image, it's not work, which includes a pr #25109 supports MTP running with cudagraph @LucasWilkinson , but may not the pr caused the error

[xiaohajiayou]

I have a proposed fix that respects EAGLE’s self.use_cuda_graph setting in dummy_run. Could you please take a look when you have time and verify if it resolves the issue?
##Root cause:

• When using PIECEWISE capture, gpu_model_runner passes use_cudagraphs=True, but the MTP heads of some draft models (e.g., DeepSeek MTP) do not natively support CUDA graphs.
• EAGLE accordingly sets self.use_cuda_graph to False and initializes self.cudagraph_batch_sizes as an empty list.
• However, PR [Spec-Decode] Support piecewise cudagraphs for Eagle head #25109 ignores EAGLE’s existing self.use_cuda_graph flag when passing arguments in dummy_run, overlooking cases where MTP heads do not support CUDA graphs, which ultimately leads to an out-of-range access in dummy_run.

[RekingQin]

I have a proposed fix that respects EAGLE’s self.use_cuda_graph setting in dummy_run. Could you please take a look when you have time and verify if it resolves the issue? ##Root cause:

• When using PIECEWISE capture, gpu_model_runner passes use_cudagraphs=True, but the MTP heads of some draft models (e.g., DeepSeek MTP) do not natively support CUDA graphs.
• EAGLE accordingly sets self.use_cuda_graph to False and initializes self.cudagraph_batch_sizes as an empty list.
• However, PR [Spec-Decode] Support piecewise cudagraphs for Eagle head #25109 ignores EAGLE’s existing self.use_cuda_graph flag when passing arguments in dummy_run, overlooking cases where MTP heads do not support CUDA graphs, which ultimately leads to an out-of-range access in dummy_run.

I have verified your fix, it fix the first error

if use_cudagraphs and num_tokens <= self.cudagraph_batch_sizes[-1]:
IndexError: list index out of range

but the second error still exists, like this

(APIServer pid=21440) INFO 10-14 16:46:25 [launcher.py:46] Route: /metrics, Methods: GET
(APIServer pid=21440) INFO:     Started server process [21440]
(APIServer pid=21440) INFO:     Waiting for application startup.
(APIServer pid=21440) INFO:     Application startup complete.
(APIServer pid=21440) INFO 10-14 16:46:33 [chat_utils.py:545] Detected the chat template content format to be 'string'. You can set `--chat-template-content-format` to override this.
(APIServer pid=21440) INFO:     127.0.0.1:33804 - "POST /v1/chat/completions HTTP/1.1" 200 OK
(Worker_TP4 pid=21975) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP6 pid=21977) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP5 pid=21976) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP7 pid=21978) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP3 pid=21974) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP0 pid=21971) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP1 pid=21972) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP2 pid=21973) WARNING 10-14 16:46:36 [flashmla_sparse.py:421] padding num_heads to 64                     due to sparse attn kernel requirement
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694] WorkerProc hit an exception.
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694] Traceback (most recent call last):
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/multiproc_executor.py", line 689, in worker_busy_loop
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     output = func(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]              ^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/worker_base.py", line 373, in execute_model
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self.worker.execute_model(scheduler_output, *args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return func(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_worker.py", line 476, in execute_model
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     output = self.model_runner.execute_model(scheduler_output, intermediate_tensors)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return func(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 2620, in execute_model
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     propose_draft_token_ids(sampler_output.sampled_token_ids)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 2585, in propose_draft_token_ids
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     self._draft_token_ids = self.propose_draft_token_ids(
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 2821, in propose_draft_token_ids
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     draft_token_ids = self.drafter.propose(
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]                       ^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/spec_decode/eagle.py", line 295, in propose
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     ret_hidden_states = self.model(
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]                         ^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/compilation/decorators.py", line 413, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     model_output = self.forward(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/deepseek_mtp.py", line 178, in forward
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     def forward(
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 929, in _fn
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return fn(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/compilation/caching.py", line 52, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self.optimized_call(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 848, in call_wrapped
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self._wrapped_call(self, *args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 424, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     raise e
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 411, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return super(self.cls, obj).__call__(*args, **kwargs)  # type: ignore[misc]
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self._call_impl(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1784, in _call_impl
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return forward_call(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "<eval_with_key>.142", line 45, in forward
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     submod_3 = self.submod_3(getitem_4, s72, getitem_5, getitem_6, submod_2);  getitem_4 = getitem_5 = getitem_6 = submod_3 = None
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 848, in call_wrapped
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self._wrapped_call(self, *args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 424, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     raise e
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/fx/graph_module.py", line 411, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return super(self.cls, obj).__call__(*args, **kwargs)  # type: ignore[misc]
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self._call_impl(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1784, in _call_impl
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return forward_call(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "<eval_with_key>.140", line 5, in forward
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     unified_mla_attention_with_output = torch.ops.vllm.unified_mla_attention_with_output(q, x_19, key_rot_1, output_8, 'model.layers.5.self_attn.attn');  q = x_19 = key_rot_1 = output_8 = unified_mla_attention_with_output = None
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]                                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/torch/_ops.py", line 1243, in __call__
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     return self._op(*args, **kwargs)
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]            ^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/attention/layer.py", line 991, in unified_mla_attention_with_output
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     self.impl.forward(
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/attention/backends/mla/flashmla_sparse.py", line 509, in forward
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     attn_metadata.req_id_per_token,
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694]     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1 pid=21972) ERROR 10-14 16:46:36 [multiproc_executor.py:694] AttributeError: 'DeepseekV32IndexerMetadata' object has no attribute 'req_id_per_token'

[xiaohajiayou]

Did you revert the change at "vllm/config/speculative.py+241" ?
We need to keep the configuration self.enforce_eager = True , because DeepSeek V3.2 does not seem to support CUDA graphs.

[RekingQin]

Did you revert the change at "vllm/config/speculative.py+241" ? We need to keep the configuration self.enforce_eager = True , because DeepSeek V3.2 does not seem to support CUDA graphs.

yeah, I reverted it

[xiaohajiayou]

I’ve identified the remaining "DeepseekV32IndexerMetadata has no attribute 'req_id_per_token'" error stems from unfiltered Indexer layers in attention lists (after PR #25103), and propose filtering them via draft_attn_layer_names - draft_indexer_layer_names in eagle.py's load_model—could you help verify if this fixes the issue?

[RekingQin]

I’ve identified the remaining "DeepseekV32IndexerMetadata has no attribute 'req_id_per_token'" error stems from unfiltered Indexer layers in attention lists (after PR #25103), and propose filtering them via draft_attn_layer_names - draft_indexer_layer_names in eagle.py's load_model—could you help verify if this fixes the issue?

yeah, it test ok right now, thank you for your effort. 😄