ml-explore/mlx

[Feature Request] Fallback for custom kernels

Fermée

#3 240 ouverte le 11 mars 2026

 (4 commentaires) (0 réaction) (0 personne assignée)C++ (1 794 forks)batch import
enhancementgood first issue

Métriques du dépôt

Stars
 (26 216 étoiles)
Métriques de merge PR
 (Merge moyen 6j 17h) (75 PRs mergées en 30 j)

Description

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.

Guide contributeur