[Bug]: Fail to load auto_round quantization format with quantized lm_head

[n1ck-guo]

Your current environment

The output of python collect_env.py

==============================
        System Info
==============================
OS                           : Ubuntu 20.04.6 LTS (x86_64)
GCC version                  : (Ubuntu 10.5.0-1ubuntu1~20.04) 10.5.0
Clang version                : Could not collect
CMake version                : version 3.22.2
Libc version                 : glibc-2.31

==============================
       PyTorch Info
==============================
PyTorch version              : 2.7.1+cu126
Is debug build               : False
CUDA used to build PyTorch   : 12.6
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.11.0 | packaged by conda-forge | (main, Jan 14 2023, 12:27:40) [GCC 11.3.0] (64-bit runtime)
Python platform              : Linux-5.4.0-169-generic-x86_64-with-glibc2.31

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 12.5.40
CUDA_MODULE_LOADING set to   : LAZY
GPU models and configuration : 
GPU 0: NVIDIA A100-SXM4-80GB
GPU 1: NVIDIA A100-SXM4-80GB
GPU 2: NVIDIA A100-SXM4-80GB
GPU 3: NVIDIA A100-SXM4-80GB
GPU 4: NVIDIA A100-SXM4-80GB
GPU 5: NVIDIA A100-SXM4-80GB
GPU 6: NVIDIA A100-SXM4-80GB
GPU 7: NVIDIA A100-SXM4-80GB

Nvidia driver version        : 555.42.02
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
Byte Order:                         Little Endian
Address sizes:                      43 bits physical, 48 bits virtual
CPU(s):                             256
On-line CPU(s) list:                0-255
Thread(s) per core:                 2
Core(s) per socket:                 64
Socket(s):                          2
NUMA node(s):                       2
Vendor ID:                          AuthenticAMD
CPU family:                         23
Model:                              49
Model name:                         AMD EPYC 7742 64-Core Processor
Stepping:                           0
Frequency boost:                    enabled
CPU MHz:                            1511.010
CPU max MHz:                        2250.0000
CPU min MHz:                        1500.0000
BogoMIPS:                           4499.79
Virtualization:                     AMD-V
L1d cache:                          4 MiB
L1i cache:                          4 MiB
L2 cache:                           64 MiB
L3 cache:                           512 MiB
NUMA node0 CPU(s):                  0-63,128-191
NUMA node1 CPU(s):                  64-127,192-255
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit:        Not affected
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Not affected
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Not affected
Vulnerability Retbleed:             Vulnerable
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 conditional, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected
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 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 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip rdpid overflow_recov succor smca sme sev sev_es

==============================
Versions of relevant libraries
==============================
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.6.4.1
[pip3] nvidia-cuda-cupti-cu12==12.6.80
[pip3] nvidia-cuda-nvrtc-cu12==12.6.77
[pip3] nvidia-cuda-runtime-cu12==12.6.77
[pip3] nvidia-cudnn-cu12==9.5.1.17
[pip3] nvidia-cufft-cu12==11.3.0.4
[pip3] nvidia-cufile-cu12==1.11.1.6
[pip3] nvidia-curand-cu12==10.3.7.77
[pip3] nvidia-cusolver-cu12==11.7.1.2
[pip3] nvidia-cusparse-cu12==12.5.4.2
[pip3] nvidia-cusparselt-cu12==0.6.3
[pip3] nvidia-nccl-cu12==2.26.2
[pip3] nvidia-nvjitlink-cu12==12.6.85
[pip3] nvidia-nvtx-cu12==12.6.77
[pip3] pyzmq==27.0.1
[pip3] torch==2.7.1
[pip3] torchaudio==2.7.1
[pip3] torchvision==0.22.1
[pip3] transformers==4.55.2
[pip3] transformers-v4.55.0-GLM-4.5V-preview==4.56.0.dev0
[pip3] triton==3.3.1
[conda] numpy                     2.2.6                    pypi_0    pypi
[conda] nvidia-cublas-cu12        12.6.4.1                 pypi_0    pypi
[conda] nvidia-cuda-cupti-cu12    12.6.80                  pypi_0    pypi
[conda] nvidia-cuda-nvrtc-cu12    12.6.77                  pypi_0    pypi
[conda] nvidia-cuda-runtime-cu12  12.6.77                  pypi_0    pypi
[conda] nvidia-cudnn-cu12         9.5.1.17                 pypi_0    pypi
[conda] nvidia-cufft-cu12         11.3.0.4                 pypi_0    pypi
[conda] nvidia-cufile-cu12        1.11.1.6                 pypi_0    pypi
[conda] nvidia-curand-cu12        10.3.7.77                pypi_0    pypi
[conda] nvidia-cusolver-cu12      11.7.1.2                 pypi_0    pypi
[conda] nvidia-cusparse-cu12      12.5.4.2                 pypi_0    pypi
[conda] nvidia-cusparselt-cu12    0.6.3                    pypi_0    pypi
[conda] nvidia-nccl-cu12          2.26.2                   pypi_0    pypi
[conda] nvidia-nvjitlink-cu12     12.6.85                  pypi_0    pypi
[conda] nvidia-nvtx-cu12          12.6.77                  pypi_0    pypi
[conda] pyzmq                     27.0.1                   pypi_0    pypi
[conda] torch                     2.7.1                    pypi_0    pypi
[conda] torchaudio                2.7.1                    pypi_0    pypi
[conda] torchvision               0.22.1                   pypi_0    pypi
[conda] transformers              4.55.2                   pypi_0    pypi
[conda] transformers-v4-55-0-glm-4-5v-preview 4.56.0.dev0              pypi_0    pypi
[conda] triton                    3.3.1                    pypi_0    pypi

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
Neuron SDK Version           : N/A
vLLM Version                 : 0.10.1.dev600+g6807af8f (git sha: 6807af8f)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-63,128-191    0               N/A
GPU1    NV12     X      NV12    NV12    NV12    NV12    NV12    NV12    0-63,128-191    0               N/A
GPU2    NV12    NV12     X      NV12    NV12    NV12    NV12    NV12    0-63,128-191    0               N/A
GPU3    NV12    NV12    NV12     X      NV12    NV12    NV12    NV12    0-63,128-191    0               N/A
GPU4    NV12    NV12    NV12    NV12     X      NV12    NV12    NV12    64-127,192-255  1               N/A
GPU5    NV12    NV12    NV12    NV12    NV12     X      NV12    NV12    64-127,192-255  1               N/A
GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X      NV12    64-127,192-255  1               N/A
GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X      64-127,192-255  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
==============================
LD_LIBRARY_PATH=/usr/local/cuda/lib64:
CUDA_HOME=/usr/local/cuda
CUDA_HOME=/usr/local/cuda
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

when loading the auto_round format quantized model with quantized lm_head. It shows error: ValueError: There is no module or parameter named 'lm_head.qweight' is Glm4MoeForCaulsalLM.

Then main problem is that the argument "prefix" in vllm/model_executor/layers/quantization/auto_round.py got empty for some layers. I add a logger to print the prefix and got the follow logs:

�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=model.layers.91.self_attn.o_proj
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:280] [model.layers.91.self_attn.o_proj] Type: RowParallelLinear, Bits: 8, Group Size: 128, Sym: True
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=model.layers.91.self_attn.attn
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:280] [model.layers.91.self_attn.attn] Type: Attention, Bits: 4, Group Size: 128, Sym: True
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=model.layers.91.mlp.experts
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:280] [model.layers.91.mlp.experts] Type: FusedMoE, Bits: 4, Group Size: 128, Sym: True
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=model.layers.91.mlp.shared_experts.gate_up_proj
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:280] [model.layers.91.mlp.shared_experts.gate_up_proj] Type: MergedColumnParallelLinear, Bits: 4, Group Size: 128, Sym: True
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=model.layers.91.mlp.shared_experts.down_proj
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:280] [model.layers.91.mlp.shared_experts.down_proj] Type: RowParallelLinear, Bits: 4, Group Size: 128, Sym: True
�[1;36m(VllmWorker TP0 pid=2079252)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [backends.py:39] Using InductorAdaptor
�[1;36m(VllmWorker TP0 pid=2079252)�[0;0m DEBUG 08-14 23:24:25 [init.py:3816] enabled custom ops: Counter()
�[1;36m(VllmWorker TP0 pid=2079252)�[0;0m DEBUG 08-14 23:24:25 [init.py:3818] disabled custom ops: Counter({'rms_norm': 369, 'silu_and_mul': 92, 'rotary_embedding': 1})
�[1;36m(VllmWorker TP0 pid=2079252)�[0;0m DEBUG 08-14 23:24:25 [base_loader.py:47] Loading weights on cuda ...
�[1;36m(VllmWorker TP2 pid=2079254)�[0;0m DEBUG 08-14 23:24:25 [auto_round.py:381] prefix=

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.

[ZJY0516]

Can you provide the command that triggered the error?

[onestardao]

This matches Problem #12 – Prefix Loss in Layer Binding from the semantic failure list.
The auto_round loader path is dropping the layer prefix before it reaches the quantized lm_head binding stage. That makes the loader unable to resolve the correct module name (GlmMoeForCausalLM.lm_head) and triggers the No module or parameter error.

It’s a semantic firewall in the argument propagation chain — the gate that should forward the prefix to each downstream layer is skipped or overwritten for some layers, so they initialize without the expected name context. A math-layer patch can restore prefix propagation without any infra change. We’ve solved similar prefix-drop cases under MIT license if you’d like a minimal reproducible fix.

[github-actions]

This issue has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this issue should remain open. Thank you!