[Bug] requirements for the version of torch and triton
awaiting response
### Checklist
- [x] 1. I have searched related issues but cannot get the expected help.
- [x] 2. The bug has not been fixed in the latest version.
- [x] 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
请问下，lmdeploy==0.12.3对torch和triton的版本有什么要求吗？
我用的是python=3.10.13，torch==2.6.0，triton==3.2.0报了这个错误
ray::RayWorkerWrapper.warmup() (pid=2431, ip=10.62.192.74, actor_id=bf0acbb5f97bd996cbd8448b01000000, repr=<lmdeploy.pytorch.engine.executor.ray_executor.RayWorkerWrapper object at 0x7f8235cdd2d0>)
File "/usr/local/lib/python3.10/concurrent/futures/_base.py", line 451, in result
return self.__get_result()
File "/usr/local/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
raise self._exception
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/engine/executor/base_worker.py", line 120, in warmup
self.model_agent.warmup()
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/engine/model_agent/agent.py", line 485, in warmup
self._forward_impl(inputs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/engine/model_agent/agent.py", line 1132, in _forward_impl
output = model_forward(
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/engine/model_agent/agent.py", line 184, in model_forward
output = model(**input_dict)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/backends/cuda/graph_runner.py", line 235, in __call__
output = self.model(**kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/models/qwen3_5.py", line 1143, in forward
hidden_states, target_inputs_embeds = self.model(
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/models/qwen3_5.py", line 1059, in forward
hidden_states = self.language_model(
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/models/qwen3_5.py", line 975, in forward
hidden_states, residual = decoder_layer(
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/models/qwen3_5.py", line 804, in forward
hidden_states = self.mlp(hidden_states, all_routed_experts=all_routed_experts)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/models/qwen3_5_moe.py", line 113, in forward
out_states = self.experts(
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/nn/moe/base.py", line 317, in forward
return self.forward_default(hidden_states, topk_weights, topk_idx)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/nn/moe/base.py", line 308, in forward_default
gemm_state = self.gemm(recv_state)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/nn/moe/default.py", line 302, in gemm
hidden_states = self.impl.forward(hidden_states,
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/backends/cuda/moe/default.py", line 58, in forward
return fused_moe(hidden_states,
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/kernels/cuda/fused_moe.py", line 598, in fused_moe
ret = moe_reduce(intermediate_cache2, topk_weights)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/lmdeploy/pytorch/kernels/cuda/fused_moe.py", line 504, in moe_reduce
_moe_reduce_kernel[grid](
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 330, in <lambda>
return lambda *args, **kwargs: self.run(grid=grid, warmup=False, *args, **kwargs)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 626, in run
kernel = self.compile(
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/triton/compiler/compiler.py", line 273, in compile
module = src.make_ir(options, codegen_fns, module_map, context)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/triton/compiler/compiler.py", line 100, in make_ir
return ast_to_ttir(self.fn, self, context=context, options=options, codegen_fns=codegen_fns,
triton.compiler.errors.CompilationError: at 43:8:
h = tl.load(h_ptrs, mask=mask_h, other=0.0)
w = tl.load(weights_ptrs, mask=mask_k, other=0.0)
if fp32_acc:
h = h.to(tl.float32)
w = w.to(tl.float32)
else:
w = w.to(h.dtype)
wh = h * w[:, None]
o = wh.sum(axis=0)
^
2026-05-06 11:03:10,805 ERROR worker.py:438 -- Unhandled error (suppress with 'RAY_IGNORE_UNHANDLED_ERRORS=1'): ray::RayWorkerWrapper.warmup() (pid=2432, ip=10.62.192.74, actor_id=96abbf61a6d8a7d8a183ae6301000000, repr=<lmdeploy.pytorch.engine.executor.ray_executor.RayWorkerWrapper object at 0x7fdafc5bd330>)
File "/home/yexun.zhang/code/mlm-data/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 770, in __call__
raise RuntimeError("Cannot call @triton.jit'd outside of the scope of a kernel")
RuntimeError: Cannot call @triton.jit'd outside of the scope of a kernel
### Reproduction
lmdeploy serve api_server /home/yexun.zhang/pretrained_models/Qwen3.5-122B-A10B --server-port 23333 --tp 8 --cache-max-entry-count 0.75 --backend pytorch
### Environment
```Shell
python：3.10.13
torch：2.6.0
triton: 3.2.0
sdk版本：1.7.0-84b81c
cuda：Cuda compilation tools, release 12.4, V12.4.1
lmdeploy: 0.12.3
命令：lmdeploy serve api_server /home/yexun.zhang/pretrained_models/Qwen3.5-122B-A10B --server-port 23333 --tp 8 --cache-max-entry-count 0.75 --backend pytorch
模型：Qwen3.5-122B-A10B
```
### Error traceback
```Shell
```
2 条评论