Description
Checklist
- 1. I have searched related issues but cannot get the expected help.
- 2. The bug has not been fixed in the latest version.
- 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
Describe the bug
I use docker from aliyun, and i run llm such as Qwen2.5-VL-7B-Instruct is OK, But when i switch to AWQ series, It's failed.
My command:
lmdeploy serve api_server
--backend turbomind
--device ascend
--eager-mode
--server-port 12000
--tp 4
--max-batch-size 32
--cache-max-entry-count 0.6
--cache-block-seq-len 64
--model-format awq
/root/Qwen2.5-VL-7B-Instruct
Error Informations:
2025-05-07 07:55:31,127 - lmdeploy - ERROR - model_agent.py:391 - Task failed
Traceback (most recent call last):
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 386, in _on_finish_callback
task.result()
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 374, in _async_loop_background
await self._async_step_background(
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 322, in _async_step_background
output = await self._async_model_forward(inputs,
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 243, in _async_model_forward
ret = await __forward(inputs)
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 220, in __forward
return await self.async_forward(inputs, swap_in_map=swap_in_map, swap_out_map=swap_out_map)
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 538, in async_forward
output = self._forward_impl(inputs, swap_in_map=swap_in_map, swap_out_map=swap_out_map)
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 521, in _forward_impl
output = model_forward(
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/opt/lmdeploy/lmdeploy/pytorch/engine/model_agent.py", line 75, in model_forward
output = model(**input_dict)
File "/opt/lmdeploy/lmdeploy/pytorch/backends/graph_runner.py", line 24, in call
return self.model(**kwargs)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/lmdeploy/lmdeploy/pytorch/models/qwen2_5_vl.py", line 439, in forward
hidden_states = self.model(
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/lmdeploy/lmdeploy/pytorch/models/qwen2_vl.py", line 295, in forward
hidden_states, residual = decoder_layer(
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/lmdeploy/lmdeploy/pytorch/models/qwen2_vl.py", line 214, in forward
hidden_states = self.self_attn(
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/lmdeploy/lmdeploy/pytorch/models/qwen2_vl.py", line 96, in forward
qkv_states = self.qkv_proj(hidden_states)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/opt/lmdeploy/lmdeploy/pytorch/nn/linear.py", line 512, in forward
out = self.impl.forward(x, self.qweight, self.scales, self.qzeros, self.bias, False)
File "/opt/lmdeploy/lmdeploy/pytorch/backends/dlinfer/awq_modules.py", line 28, in forward
out = awq_linear(x, qweight, scales, qzeros, bias, all_reduce, self.group_size)
File "/opt/lmdeploy/lmdeploy/pytorch/kernels/dlinfer/awq_kernels.py", line 15, in awq_linear
return ext_ops.weight_quant_matmul(x.squeeze(0),
File "/usr/local/python3.10.5/lib/python3.10/site-packages/dlinfer/ops/llm.py", line 542, in weight_quant_matmul
return vendor_ops_registry["weight_quant_matmul"](
File "/usr/local/python3.10.5/lib/python3.10/site-packages/dlinfer/vendor/ascend/torch_npu_ops.py", line 404, in weight_quant_matmul
return torch.ops.npu.npu_weight_quant_batchmatmul(
File "/usr/local/python3.10.5/lib/python3.10/site-packages/torch/ops.py", line 854, in call
return self._op(*args, **(kwargs or {}))
RuntimeError: call aclnnWeightQuantBatchMatmulV2 failed, detail:EZ1001: [PID: 38901] 2025-05-07-07:55:31.121.354 antiquantScale's dtype must be DT_UINT64 or DT_INT64 when antiquantOffset's dtype is DT_INT32, actual antiquantScale's dtype is [DT_FLOAT16].
Reproduction
lmdeploy serve api_server
--backend turbomind
--device ascend
--eager-mode
--server-port 12000
--tp 4
--max-batch-size 32
--cache-max-entry-count 0.6
--cache-block-seq-len 64
--model-format awq
/root/Qwen2.5-VL-7B-Instruct
Environment
lmdeploy check_env
[W compiler_depend.ts:615] Warning: expandable_segments currently defaults to false. You can enable this feature by `export PYTORCH_NPU_ALLOC_CONF = expandable_segments:True`. (function operator())
sys.platform: linux
Python: 3.10.5 (main, Mar 24 2025, 07:28:13) [GCC 9.4.0]
CUDA available: False
MUSA available: False
numpy_random_seed: 2147483648
GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
PyTorch: 2.3.1
PyTorch compiling details: PyTorch built with:
- GCC 10.2
- C++ Version: 201703
- Intel(R) MKL-DNN v3.3.6 (Git Hash 86e6af5974177e513fd3fee58425e1063e7f1361)
- OpenMP 201511 (a.k.a. OpenMP 4.5)
- LAPACK is enabled (usually provided by MKL)
- NNPACK is enabled
- CPU capability usage: NO AVX
- Build settings: BLAS_INFO=open, BUILD_TYPE=Release, CXX_COMPILER=/opt/rh/devtoolset-10/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOCUPTI -DLIBKINETO_NOROCTRACER -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wsuggest-override -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-missing-braces -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=open, TORCH_VERSION=2.3.1, USE_CUDA=OFF, USE_CUDNN=OFF, USE_CUSPARSELT=OFF, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_GLOO=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=OFF, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, USE_ROCM_KERNEL_ASSERT=OFF,
TorchVision: 0.18.1
LMDeploy: 0.7.3+
transformers: 4.51.3
gradio: Not Found
fastapi: 0.115.12
pydantic: 2.11.3
triton: Not Found