具体支持哪些npu设备呢？

[qiaonanxiansheng]

(PaddleOCR-NPU) [root@localhost PaddleOCR-NPU]# python quick_ocr.py --image_dir doc/imgs/ --det_model_path ./models/ptocr_v5_server_det.pth --rec_model_path ./models/ptocr_v5_server_rec.pth --cls_model_path ./models/ch_ptocr_mobile_v2.0_cls_infer.pth

🔥 Quick OCR - 简化的OCR命令行工具

🚀 初始化OCR实例...
Warning : ASCEND_HOME_PATH environment variable is not set.
使用华为NPU设备: npu:0
Warning: since the loaded file is not a zipfile, only "torch.device" and "str" type parameters are currently supported for parameter types of map_locationIf parameter types of map_location is "Callable[[torch.Tensor, str], torch.Tensor]" or "Dict[str, str]", which is only support for zipfile,all tensors are currently loaded onto the CPU, which may introduce problems
model is loaded: ./models/ptocr_v5_server_det.pth
PytorchPaddleOCR 初始化失败: Unsupported soc version: Ascend310
❌ OCR实例初始化失败: Unsupported soc version: Ascend310

[BytePioneer-AI]

目前来看，910A、910B肯定可以，310P系列有一些同学遇到了问题。
但你这个错误，我感觉是CANN的报错或者是torch-npu的报错，你修改CANN版本为你对应芯片的了吗，同时torch-npu也需要使用对应版本