huggingface runtime_error ai_generated partial

运行时错误：MPS设备上未分配占位符存储

RuntimeError: Placeholder storage has not been allocated on MPS device

ID: huggingface/mps-fp16-cast-error

其他格式: JSON · Markdown 中文 · English
80%修复率
85%置信度
1证据数
2023-06-15首次发现

版本兼容性

版本状态引入弃用备注
transformers>=4.30.0 active — — —
torch>=2.0.0 active — — —
macOS>=13.0 active — — —

根因分析

模型权重或张量在Apple MPS后端上被转换为float16，但MPS不完全支持FP16，导致分配失败。

English

Model weights or tensors are being cast to float16 on Apple MPS backend which does not fully support FP16, causing allocation failure.

generic

官方文档

https://pytorch.org/docs/stable/notes/mps.html

解决方案

  1. Load model with torch_dtype=torch.float32 and use model.to('mps') without half precision.
  2. Set environment variable PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 to disable MPS memory allocation optimization.
  3. Use CPU fallback by setting device='cpu' and use float16 with CPU if memory is a concern.

无效尝试

常见但无效的做法:

  1. 95% 失败

    MPS backend has limited FP16 support; explicit FP16 casting triggers the error.

  2. 90% 失败

    Half-precision conversion on MPS is not fully implemented and causes memory allocation errors.