vllm - 💡(How to fix) Fix [Bug]: Gemma4 multimodal crashes with "pixel_values contains inconsistent shapes" when concurrent image requests have different resolutions [1 participants]

vllm2026-04-13 07:59:35

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

vllm-project/vllm#39681•Fetched 2026-04-14 05:38:10

View on GitHub

Comments

Participants

Timeline

Reactions

Author

art3na

Participants

art3na

Timeline (top)

labeled ×1

Error Message

encoder co-batches them and crashes during input validation. The error takes Full traceback (Worker_TP0) ERROR multiproc_executor.py:949 WorkerProc hit an exception. Traceback (most recent call last):

Root Cause

Root cause (brief)

Fix Action

Fix / Workaround

============================== 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 Model name: AMD EPYC 9654 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: 3709.3569 CPU min MHz: 1500.0000 BogoMIPS: 4800.30 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 rapl 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 ibrs_enhanced 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 rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d ibpb_exit_to_user 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 Gather data sampling: Not affected Vulnerability Indirect target selection: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Reg file data sampling: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Mitigation; safe RET Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected Vulnerability Srbds: Not affected Vulnerability Tsa: Mitigation; Clear CPU buffers Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Mitigation; IBPB before exit to userspace

The Gemma4 HF image processor sizes pixel_values per image as (max_patches, patch_pixels) where max_patches = max_soft_tokens × pooling_kernel_size², and max_soft_tokens depends on image resolution. Different images therefore get different max_patches (e.g. 10080 vs 2520).

Accept pixel_values as list[Tensor] of varying shape in _parse_and_validate_image_input and pass through to the existing per-image loop in _process_image_input (minimal change).
Per the existing TODO at gemma4_mm.py:1049, pad all images in a batch to a common max_patches in _call_hf_processor.

Code Example

==============================
        System Info
==============================
OS                           : Ubuntu 22.04.5 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.35

==============================
       PyTorch Info
==============================
PyTorch version              : 2.10.0+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.13 (main, Mar  4 2026, 09:23:07) [GCC 11.4.0] (64-bit runtime)
Python platform              : Linux-5.15.0-174-generic-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.0.88
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

Nvidia driver version        : 590.48.01
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
Model name:                              AMD EPYC 9654 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:                             3709.3569
CPU min MHz:                             1500.0000
BogoMIPS:                                4800.30
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 rapl 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 ibrs_enhanced 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 rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d ibpb_exit_to_user
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 Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Mitigation; safe RET
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Mitigation; Clear CPU buffers
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.6
[pip3] numpy==2.2.6
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.15.1.9
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft==12.0.0.61
[pip3] nvidia-cufile==1.15.1.6
[pip3] nvidia-curand==10.4.0.35
[pip3] nvidia-cusolver==12.0.4.66
[pip3] nvidia-cusparse==12.6.3.3
[pip3] nvidia-cusparselt-cu13==0.8.0
[pip3] nvidia-cutlass-dsl==4.4.2
[pip3] nvidia-cutlass-dsl-libs-base==4.4.2
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu13==2.28.9
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] pyzmq==27.1.0
[pip3] torch==2.10.0+cu130
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.10.0+cu130
[pip3] torchvision==0.25.0+cu130
[pip3] transformers==5.5.0
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.19.1.dev6+g6d4a8e6d2 (git sha: 6d4a8e6d2)
vLLM Build Flags:
  CUDA Archs: 7.0 7.5 8.0 8.9 9.0 10.0 12.0; ROCm: Disabled
GPU Topology:
  	GPU0	GPU1	GPU2	GPU3	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	NV18	NV18	NV18	96-191,288-383	1		N/A
GPU1	NV18	 X 	NV18	NV18	96-191,288-383	1		N/A
GPU2	NV18	NV18	 X 	NV18	96-191,288-383	1		N/A
GPU3	NV18	NV18	NV18	 X 	96-191,288-383	1		N/A

Legend:

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

==============================
     Environment Variables
==============================
NVIDIA_VISIBLE_DEVICES=all
NVIDIA_REQUIRE_CUDA=cuda>=13.0 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551 brand=unknown,driver>=565,driver<566 brand=grid,driver>=565,driver<566 brand=tesla,driver>=565,driver<566 brand=nvidia,driver>=565,driver<566 brand=quadro,driver>=565,driver<566 brand=quadrortx,driver>=565,driver<566 brand=nvidiartx,driver>=565,driver<566 brand=vapps,driver>=565,driver<566 brand=vpc,driver>=565,driver<566 brand=vcs,driver>=565,driver<566 brand=vws,driver>=565,driver<566 brand=cloudgaming,driver>=565,driver<566 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571 brand=unknown,driver>=575,driver<576 brand=grid,driver>=575,driver<576 brand=tesla,driver>=575,driver<576 brand=nvidia,driver>=575,driver<576 brand=quadro,driver>=575,driver<576 brand=quadrortx,driver>=575,driver<576 brand=nvidiartx,driver>=575,driver<576 brand=vapps,driver>=575,driver<576 brand=vpc,driver>=575,driver<576 brand=vcs,driver>=575,driver<576 brand=vws,driver>=575,driver<576 brand=cloudgaming,driver>=575,driver<576
TORCH_CUDA_ARCH_LIST=7.0 7.5 8.0 8.9 9.0 10.0 12.0
NVIDIA_DRIVER_CAPABILITIES=compute,utility
VLLM_USAGE_SOURCE=production-docker-image
CUDA_VERSION=13.0.1
VLLM_ENABLE_CUDA_COMPATIBILITY=0
LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_root

---

docker run -itd --gpus '"device=4,5,6,7"' --name vllm-gemma-4-31b-it \
  --ipc=host --network host --shm-size 16G \
  -v /data/nlp/models:/models/ \
  vllm/vllm-openai:gemma4-cu130 \
  --model /models/gemma-4-31B-it \
  --tensor-parallel-size 4 \
  --max-model-len 262144 \
  --gpu-memory-utilization 0.90 \
  --host 0.0.0.0 --port 8001 \
  --served-model-name gemma4 \
  --enable-auto-tool-choice \
  --reasoning-parser gemma4 \
  --tool-call-parser gemma4

RAW_BUFFERClick to expand / collapse

Your current environment

<details> <summary>The output of <code>python collect_env.py</code></summary>

==============================
        System Info
==============================
OS                           : Ubuntu 22.04.5 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0
Clang version                : Could not collect
CMake version                : Could not collect
Libc version                 : glibc-2.35

==============================
       PyTorch Info
==============================
PyTorch version              : 2.10.0+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.13 (main, Mar  4 2026, 09:23:07) [GCC 11.4.0] (64-bit runtime)
Python platform              : Linux-5.15.0-174-generic-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.0.88
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

Nvidia driver version        : 590.48.01
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
Model name:                              AMD EPYC 9654 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:                             3709.3569
CPU min MHz:                             1500.0000
BogoMIPS:                                4800.30
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 rapl 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 ibrs_enhanced 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 rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d ibpb_exit_to_user
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 Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Mitigation; safe RET
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Mitigation; Clear CPU buffers
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.6
[pip3] numpy==2.2.6
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.15.1.9
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft==12.0.0.61
[pip3] nvidia-cufile==1.15.1.6
[pip3] nvidia-curand==10.4.0.35
[pip3] nvidia-cusolver==12.0.4.66
[pip3] nvidia-cusparse==12.6.3.3
[pip3] nvidia-cusparselt-cu13==0.8.0
[pip3] nvidia-cutlass-dsl==4.4.2
[pip3] nvidia-cutlass-dsl-libs-base==4.4.2
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu13==2.28.9
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] pyzmq==27.1.0
[pip3] torch==2.10.0+cu130
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.10.0+cu130
[pip3] torchvision==0.25.0+cu130
[pip3] transformers==5.5.0
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.19.1.dev6+g6d4a8e6d2 (git sha: 6d4a8e6d2)
vLLM Build Flags:
  CUDA Archs: 7.0 7.5 8.0 8.9 9.0 10.0 12.0; ROCm: Disabled
GPU Topology:
  	GPU0	GPU1	GPU2	GPU3	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	NV18	NV18	NV18	96-191,288-383	1		N/A
GPU1	NV18	 X 	NV18	NV18	96-191,288-383	1		N/A
GPU2	NV18	NV18	 X 	NV18	96-191,288-383	1		N/A
GPU3	NV18	NV18	NV18	 X 	96-191,288-383	1		N/A

Legend:

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

==============================
     Environment Variables
==============================
NVIDIA_VISIBLE_DEVICES=all
NVIDIA_REQUIRE_CUDA=cuda>=13.0 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551 brand=unknown,driver>=565,driver<566 brand=grid,driver>=565,driver<566 brand=tesla,driver>=565,driver<566 brand=nvidia,driver>=565,driver<566 brand=quadro,driver>=565,driver<566 brand=quadrortx,driver>=565,driver<566 brand=nvidiartx,driver>=565,driver<566 brand=vapps,driver>=565,driver<566 brand=vpc,driver>=565,driver<566 brand=vcs,driver>=565,driver<566 brand=vws,driver>=565,driver<566 brand=cloudgaming,driver>=565,driver<566 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571 brand=unknown,driver>=575,driver<576 brand=grid,driver>=575,driver<576 brand=tesla,driver>=575,driver<576 brand=nvidia,driver>=575,driver<576 brand=quadro,driver>=575,driver<576 brand=quadrortx,driver>=575,driver<576 brand=nvidiartx,driver>=575,driver<576 brand=vapps,driver>=575,driver<576 brand=vpc,driver>=575,driver<576 brand=vcs,driver>=575,driver<576 brand=vws,driver>=575,driver<576 brand=cloudgaming,driver>=575,driver<576
TORCH_CUDA_ARCH_LIST=7.0 7.5 8.0 8.9 9.0 10.0 12.0
NVIDIA_DRIVER_CAPABILITIES=compute,utility
VLLM_USAGE_SOURCE=production-docker-image
CUDA_VERSION=13.0.1
VLLM_ENABLE_CUDA_COMPATIBILITY=0
LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_root

</details>

🐛 Describe the bug

When two or more concurrent chat-completion requests containing images of different resolutions land in the same scheduler step, Gemma4's multimodal encoder co-batches them and crashes during input validation. The error takes down all TP workers and the engine core; the server stops serving requests.

ValueError: pixel_values contains inconsistent shapes: torch.Size([10080, 768]) (index 0) vs torch.Size([2520, 768]) (index 1)

Root cause (brief)

Gemma4ImagePixelInputs declares pixel_values with a uniform ("bn", "np", "pp") shape (https://github.com/vllm-project/vllm/blob/main/vllm/model_executor/models/gemma4_mm.py#L92-L112), so when _execute_mm_encoder stacks images from different requests, the TensorSchema validator rejects the mismatched np dim. Note that _process_image_input already loops one image at a time through the vision tower — the underlying forward path tolerates variable shapes; only the schema/stacking layer enforces uniformity (the TODO at line 1049 anticipates this).

Reproduction

Server (the exact docker run I used):

docker run -itd --gpus '"device=4,5,6,7"' --name vllm-gemma-4-31b-it \
  --ipc=host --network host --shm-size 16G \
  -v /data/nlp/models:/models/ \
  vllm/vllm-openai:gemma4-cu130 \
  --model /models/gemma-4-31B-it \
  --tensor-parallel-size 4 \
  --max-model-len 262144 \
  --gpu-memory-utilization 0.90 \
  --host 0.0.0.0 --port 8001 \
  --served-model-name gemma4 \
  --enable-auto-tool-choice \
  --reasoning-parser gemma4 \
  --tool-call-parser gemma4

Client — fire two requests in parallel with images of clearly different resolutions:

import asyncio, base64, io import httpx from PIL import Image

URL = "http://localhost:8001/v1/chat/completions"

def b64_image(w, h, color): buf = io.BytesIO() Image.new("RGB", (w, h), color).save(buf, format="PNG") return "data:image/png;base64," + base64.b64encode(buf.getvalue()).decode()

def payload(img_url): return { "model": "gemma4", "messages": [{ "role": "user", "content": [ {"type": "image_url", "image_url": {"url": img_url}}, {"type": "text", "text": "Describe this image."}, ], }], "max_tokens": 32, }

async def main(): big = b64_image(1024, 1024, "red") small = b64_image(256, 128, "blue") async with httpx.AsyncClient(timeout=120) as c: r1, r2 = await asyncio.gather( c.post(URL, json=payload(big)), c.post(URL, json=payload(small)), ) print(r1.status_code, r1.text[:200]) print(r2.status_code, r2.text[:200])

asyncio.run(main()) The two requests get co-batched in the MM encoder and the engine crashes deterministically (within a few attempts, depending on scheduler timing — repeat the script in a loop if needed).

Full traceback

(Worker_TP0) ERROR multiproc_executor.py:949 WorkerProc hit an exception. Traceback (most recent call last): File ".../vllm/v1/executor/multiproc_executor.py", line 944, in worker_busy_loop output = func(*args, **kwargs) File ".../vllm/v1/worker/worker_base.py", line 332, in execute_model return self.worker.execute_model(scheduler_output) File ".../vllm/v1/worker/gpu_worker.py", line 803, in execute_model output = self.model_runner.execute_model(...) File ".../vllm/v1/worker/gpu_model_runner.py", line 3992, in execute_model ) = self._preprocess(...) File ".../vllm/v1/worker/gpu_model_runner.py", line 3228, in _preprocess self._execute_mm_encoder(scheduler_output) File ".../vllm/v1/worker/gpu_model_runner.py", line 2886, in _execute_mm_encoder batch_outputs = model.embed_multimodal(**mm_kwargs_batch) File ".../vllm/model_executor/models/gemma4_mm.py", line 1188, in embed_multimodal mm_input_by_modality = self._parse_and_validate_multimodal_inputs(**kwargs) File ".../vllm/model_executor/models/gemma4_mm.py", line 1003, in _parse_and_validate_multimodal_inputs mm_input_by_modality["image"] = self._parse_and_validate_image_input(...) File ".../vllm/model_executor/models/gemma4_mm.py", line 961, in _parse_and_validate_image_input return Gemma4ImagePixelInputs(...) File ".../vllm/utils/tensor_schema.py", line 63, in init self.validate() File ".../vllm/utils/tensor_schema.py", line 230, in validate actual_shape = self._validate_field(...) File ".../vllm/utils/tensor_schema.py", line 145, in _validate_field raise ValueError(...) ValueError: pixel_values contains inconsistent shapes: torch.Size([10080, 768]) (index 0) vs torch.Size([2520, 768]) (index 1) (All four Worker_TP0..3 processes raise the same ValueError, then the engine core dies.)

Suggested fix Either:

Accept pixel_values as list[Tensor] of varying shape in _parse_and_validate_image_input and pass through to the existing per-image loop in _process_image_input (minimal change).
Per the existing TODO at gemma4_mm.py:1049, pad all images in a batch to a common max_patches in _call_hf_processor.

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.

extent analysis

TL;DR

The most likely fix is to modify the _parse_and_validate_image_input function to accept pixel_values as a list of tensors with varying shapes and pass them through to the existing per-image loop in _process_image_input.

Guidance

Identify the root cause of the issue: The error occurs because the Gemma4ImagePixelInputs class expects a uniform shape for pixel_values, but the max_patches dimension varies depending on the image resolution.
Consider the two suggested fixes:
- Accept pixel_values as a list of tensors with varying shapes in _parse_and_validate_image_input.
- Pad all images in a batch to a common max_patches in _call_hf_processor.
Evaluate the trade-offs between these two approaches, considering factors such as performance, complexity, and maintainability.

Example

No code example is provided, as the issue requires a modification to the existing codebase, and the suggested fixes are described in the issue text.

Notes

The issue is specific to the Gemma4 model and the Gemma4ImagePixelInputs class. The suggested fixes may have implications for performance, memory usage, or other aspects of the system, and should be thoroughly tested before implementation.

Recommendation

Apply the first suggested fix: accept pixel_values as a list of tensors with varying shapes in _parse_and_validate_image_input. This approach seems to be the most minimal and straightforward solution, and it aligns with the existing per-image loop in _process_image_input.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #ssr #environment variable #API versioning #request timeout

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

vllm - 💡(How to fix) Fix [Bug]: Gemma4 multimodal crashes with "pixel_values contains inconsistent shapes" when concurrent image requests have different resolutions [1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Root cause (brief)

Fix Action

Fix / Workaround

============================== CPU Info

Code Example

Your current environment

🐛 Describe the bug

Root cause (brief)

Reproduction

Before submitting a new issue...

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

TRENDING

vllm - 💡(How to fix) Fix [Bug]: Gemma4 multimodal crashes with "pixel_values contains inconsistent shapes" when concurrent image requests have different resolutions [1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Root cause (brief)

Fix Action

Fix / Workaround

============================== CPU Info

Code Example

Your current environment

🐛 Describe the bug

Root cause (brief)

Reproduction

Before submitting a new issue...

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

RELATED_DISCOVERY

TRENDING