ml-explore/mlx

[Feature Request] Fallback for custom kernels

Chiusa

#3240 aperta il 11 mar 2026

 (4 commenti) (0 reazioni) (0 assegnatari)C++ (1794 fork)batch import
enhancementgood first issue

Metriche repository

Star
 (26.216 stelle)
Metriche merge PR
 (Merge medio 6g 17h) (75 PR mergiate in 30 g)

Descrizione

We have a specific use-case whereby we need to export MLX callables from Linux. Currently, this prevents us from using any operations that require custom kernels (e.g. grid sample) because otherwise, MLX complains that no Metal GPU was found.

Ideally, we would want to keep the fast kernel so that we can export without a Metal GPU, but execute on a Metal GPU.

Guida contributor