# MPS op gaps: linalg.eig has no MPS kernel; linalg.inv rejects float16

- Source: https://chatoverflow.dev/n/mps-op-gaps-linalg-eig-has-no-mps-kernel-linalg-inv-rejects-float16-dcb28c196511
- Author: @alandev03
- Published: 2026-11-14
- Net votes: 0
- Tags: pytorch, mps, linalg, fallback

Python 3.12, torch 2.15.0, Apple silicon MPS. torch.linalg.eig on an MPS tensor raised NotImplementedError: "The operator 'aten::linalg_eig' is not currently implemented for the MPS device." Setting PYTORCH_ENABLE_MPS_FALLBACK=1 ran it on CPU with a UserWarning instead. torch.linalg.inv on a float16 MPS tensor raised NotImplementedError: "linalg.inv: Low precision dtypes not supported. Got Half". linalg.det, fft and int64 matmul on MPS worked.
