[Bug]: GPTQ cannot be used on ROCm after upgraded vLLM to v0.8.5

[nlzy]

Your current environment

The output of `python collect_env.py`

INFO 04-29 19:57:29 [__init__.py:239] Automatically detected platform rocm.
Collecting environment information...
PyTorch version: 2.7.0+rocm6.3
Is debug build: False
CUDA used to build PyTorch: N/A
ROCM used to build PyTorch: 6.3.42131-fa1d09cbd

OS: Ubuntu 24.04.2 LTS (x86_64)
GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
Clang version: Could not collect
CMake version: version 3.31.6
Libc version: glibc-2.39

Python version: 3.12.3 (main, Feb  4 2025, 14:48:35) [GCC 13.3.0] (64-bit runtime)
Python platform: Linux-6.8.12-8-pve-x86_64-with-glibc2.39
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: AMD Radeon Graphics (gfx906:sramecc+:xnack-)
Nvidia driver version: Could not collect
cuDNN version: Could not collect
HIP runtime version: 6.3.42131
MIOpen runtime version: 3.3.0
Is XNNPACK available: True

CPU:
Architecture:                         x86_64
CPU op-mode(s):                       32-bit, 64-bit
Address sizes:                        46 bits physical, 48 bits virtual
Byte Order:                           Little Endian
CPU(s):                               20
On-line CPU(s) list:                  0-19
Vendor ID:                            GenuineIntel
Model name:                           Intel(R) Xeon(R) CPU E5-2666 v3 @ 2.90GHz
CPU family:                           6
Model:                                63
Thread(s) per core:                   2
Core(s) per socket:                   10
Socket(s):                            1
Stepping:                             2
CPU(s) scaling MHz:                   93%
CPU max MHz:                          3500.0000
CPU min MHz:                          1200.0000
BogoMIPS:                             5786.34
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb pti intel_ppin ssbd ibrs ibpb stibp tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm xsaveopt cqm_llc cqm_occup_llc dtherm ida arat pln pts vnmi md_clear flush_l1d
Virtualization:                       VT-x
L1d cache:                            320 KiB (10 instances)
L1i cache:                            320 KiB (10 instances)
L2 cache:                             2.5 MiB (10 instances)
L3 cache:                             25 MiB (1 instance)
NUMA node(s):                         1
NUMA node0 CPU(s):                    0-19
Vulnerability Gather data sampling:   Not affected
Vulnerability Itlb multihit:          KVM: Mitigation: VMX disabled
Vulnerability L1tf:                   Mitigation; PTE Inversion; VMX conditional cache flushes, SMT vulnerable
Vulnerability Mds:                    Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Meltdown:               Mitigation; PTI
Vulnerability Mmio stale data:        Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed:               Not affected
Vulnerability Spec rstack overflow:   Not affected
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:             Mitigation; Retpolines; IBPB conditional; IBRS_FW; STIBP conditional; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] numpy==2.1.2
[pip3] pytorch-triton-rocm==3.3.0
[pip3] pyzmq==26.4.0
[pip3] torch==2.7.0+rocm6.3
[pip3] torchaudio==2.7.0+rocm6.3
[pip3] torchvision==0.22.0+rocm6.3
[pip3] transformers==4.51.3
[pip3] triton==3.3.0+git9c06a19c
[conda] Could not collect
ROCM Version: 6.3.42134-a9a80e791
Neuron SDK Version: N/A
vLLM Version: 0.8.6.dev22+gec44935b0 (git sha: ec44935b0)
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
============================ ROCm System Management Interface ============================
================================ Weight between two GPUs =================================
       GPU0         GPU1
GPU0   0            40
GPU1   40           0

================================= Hops between two GPUs ==================================
       GPU0         GPU1
GPU0   0            2
GPU1   2            0

=============================== Link Type between two GPUs ===============================
       GPU0         GPU1
GPU0   0            PCIE
GPU1   PCIE         0

======================================= Numa Nodes =======================================
GPU[0]		: (Topology) Numa Node: 0
GPU[0]		: (Topology) Numa Affinity: -1
GPU[1]		: (Topology) Numa Node: 0
GPU[1]		: (Topology) Numa Affinity: -1
================================== End of ROCm SMI Log ===================================

PYTORCH_ROCM_ARCH=gfx906
TORCH_BLAS_PREFER_HIPBLASLT=0
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

I have been using GPTQ quantization on the ROCm platform, and I created an API server using the simple vllm serve command, which worked perfectly.

However, after updating vLLM to version v0.8.5, the vllm serve command failed to start, showing the error message: "gptq_bitblas quantization is currently not supported in rocm."

After manually adding the --quantization gptq parameter, vLLM worked normally as it did before.

It should be noted that the GPU I am using is very old and no longer supported. Therefore, the vLLM I am using is a downstream branch that I maintain myself. However, I believe this issue also exists in the upstream branch here, so I am submitting this issue here.

Below is the full log output:

(vllm085) nlzy@Nlzy-Lorazepam-100 ~/vllm ((ec44935b))> vllm serve /mnt/Qwen2.5-7B-Instruct-GPTQ-Int4/
INFO 04-29 19:56:35 [__init__.py:239] Automatically detected platform rocm.
INFO 04-29 19:56:44 [api_server.py:1043] vLLM API server version 0.8.6.dev22+gec44935b0
INFO 04-29 19:56:44 [api_server.py:1044] args: Namespace(subparser='serve', model_tag='/mnt/Qwen2.5-7B-Instruct-GPTQ-Int4/', config='', host=None, port=8000, uvicorn_log_level='info', disable_uvicorn_access_log=False, allow_credentials=False, allowed_origins=['*'], allowed_methods=['*'], allowed_headers=['*'], api_key=None, lora_modules=None, prompt_adapters=None, chat_template=None, chat_template_content_format='auto', response_role='assistant', ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, enable_ssl_refresh=False, ssl_cert_reqs=0, root_path=None, middleware=[], return_tokens_as_token_ids=False, disable_frontend_multiprocessing=False, enable_request_id_headers=False, enable_auto_tool_choice=False, tool_call_parser=None, tool_parser_plugin='', model='/mnt/Qwen2.5-7B-Instruct-GPTQ-Int4/', task='auto', tokenizer=None, hf_config_path=None, skip_tokenizer_init=False, revision=None, code_revision=None, tokenizer_revision=None, tokenizer_mode='auto', trust_remote_code=False, allowed_local_media_path=None, load_format='auto', download_dir=None, model_loader_extra_config={}, use_tqdm_on_load=True, config_format=<ConfigFormat.AUTO: 'auto'>, dtype='auto', max_model_len=None, guided_decoding_backend='auto', reasoning_parser=None, logits_processor_pattern=None, model_impl='auto', distributed_executor_backend=None, pipeline_parallel_size=1, tensor_parallel_size=1, data_parallel_size=1, enable_expert_parallel=False, max_parallel_loading_workers=None, ray_workers_use_nsight=False, disable_custom_all_reduce=False, block_size=None, gpu_memory_utilization=0.9, swap_space=4, kv_cache_dtype='auto', num_gpu_blocks_override=None, enable_prefix_caching=None, prefix_caching_hash_algo='builtin', cpu_offload_gb=0, calculate_kv_scales=False, disable_sliding_window=False, use_v2_block_manager=True, seed=None, max_logprobs=20, disable_log_stats=False, quantization=None, rope_scaling=None, rope_theta=None, hf_token=None, hf_overrides=None, enforce_eager=False, max_seq_len_to_capture=8192, tokenizer_pool_size=0, tokenizer_pool_type='ray', tokenizer_pool_extra_config={}, limit_mm_per_prompt={}, mm_processor_kwargs=None, disable_mm_preprocessor_cache=False, enable_lora=None, enable_lora_bias=False, max_loras=1, max_lora_rank=16, lora_extra_vocab_size=256, lora_dtype='auto', long_lora_scaling_factors=None, max_cpu_loras=None, fully_sharded_loras=False, enable_prompt_adapter=None, max_prompt_adapters=1, max_prompt_adapter_token=0, device='auto', speculative_config=None, ignore_patterns=[], served_model_name=None, qlora_adapter_name_or_path=None, show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, disable_async_output_proc=False, max_num_batched_tokens=None, max_num_seqs=None, max_num_partial_prefills=1, max_long_partial_prefills=1, long_prefill_token_threshold=0, num_lookahead_slots=0, scheduler_delay_factor=0.0, preemption_mode=None, num_scheduler_steps=1, multi_step_stream_outputs=True, scheduling_policy='fcfs', enable_chunked_prefill=None, disable_chunked_mm_input=False, scheduler_cls='vllm.core.scheduler.Scheduler', override_neuron_config=None, override_pooler_config=None, compilation_config=None, kv_transfer_config=None, worker_cls='auto', worker_extension_cls='', generation_config='auto', override_generation_config=None, enable_sleep_mode=False, additional_config=None, enable_reasoning=False, disable_cascade_attn=False, disable_log_requests=False, max_log_len=None, disable_fastapi_docs=False, enable_prompt_tokens_details=False, enable_server_load_tracking=False, dispatch_function=<function ServeSubcommand.cmd at 0x75590c8479c0>)
INFO 04-29 19:56:59 [config.py:717] This model supports multiple tasks: {'reward', 'classify', 'score', 'embed', 'generate'}. Defaulting to 'generate'.
INFO 04-29 19:56:59 [gptq_bitblas.py:164] The model is convertible to gptq_bitblas during runtime. Using gptq_bitblas kernel.
Traceback (most recent call last):
  File "/home/nlzy/vllm085/bin/vllm", line 8, in <module>
    sys.exit(main())
             ^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/entrypoints/cli/main.py", line 53, in main
    args.dispatch_function(args)
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/entrypoints/cli/serve.py", line 27, in cmd
    uvloop.run(run_server(args))
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/uvloop/__init__.py", line 109, in run
    return __asyncio.run(
           ^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 194, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "uvloop/loop.pyx", line 1518, in uvloop.loop.Loop.run_until_complete
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/uvloop/__init__.py", line 61, in wrapper
    return await main
           ^^^^^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/entrypoints/openai/api_server.py", line 1078, in run_server
    async with build_async_engine_client(args) as engine_client:
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/entrypoints/openai/api_server.py", line 146, in build_async_engine_client
    async with build_async_engine_client_from_engine_args(
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/entrypoints/openai/api_server.py", line 166, in build_async_engine_client_from_engine_args
    vllm_config = engine_args.create_engine_config(usage_context=usage_context)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/engine/arg_utils.py", line 1099, in create_engine_config
    model_config = self.create_model_config()
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/engine/arg_utils.py", line 987, in create_model_config
    return ModelConfig(
           ^^^^^^^^^^^^
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/config.py", line 546, in __init__
    self._verify_quantization()
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/config.py", line 828, in _verify_quantization
    current_platform.verify_quantization(self.quantization)
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/platforms/rocm.py", line 291, in verify_quantization
    super().verify_quantization(quant)
  File "/home/nlzy/vllm085/lib/python3.12/site-packages/vllm/platforms/interface.py", line 287, in verify_quantization
    raise ValueError(
ValueError: gptq_bitblas quantization is currently not supported in rocm.

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.