qwen3.5的mnn模型如何进行gpu推理

您好 在wsl环境下基于此步骤安装mnn_cuda模块1、进入pymnn/pip_package 目录 2、python3 build_deps.py llm,cuda 3、sudo python3 setup.py install --deps llm,cuda 安装完成后进行推理时总是报错 换成最小的模型0.8B也跑不了报如下错误 显卡是rtx5060 这是什么原因呢 是因为用的wsl环境吗 @DESKTOP-QV06STS:/mnt/c/Windows/system32$ python /mnt/d/MNN-master/pymnn/examples/MNNLlm/vlm_example.py /mnt/d/MNN-master/models/mnn/config.json The device supports: i8sdot:0, fp16:0, i8mm: 0, sve2: 0, sme2: 0 cublasCreate failed: 1 Cache invalid, will be reset CUDA alloc failed CUDA alloc failed Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. CUDA alloc failed CUDA alloc failed CUDA alloc failed CUDA alloc failed

......

Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. Out of memory when gamma is acquired in CudaLayerNorm. Out of memory when beta is acquired in CudaLayerNorm. code=1 in onForward, 666 Segmentation fault python /mnt/d/MNN-master/pymnn/examples/MNNLlm/vlm_example.py /mnt/d/MNN-master/models/mnn/config.json