ml-explore/mlx

[Feature Request] Fallback for custom kernels

Geschlossen

#3.240 geöffnet am 11.03.2026

 (4 Kommentare) (0 Reaktionen) (0 zugewiesene Personen)C++ (1.794 Forks)batch import
enhancementgood first issue

Repository-Metriken

Stars
 (26.216 Sterne)
PR-Merge-Metriken
 (Durchschn. Merge 6T 17h) (75 gemergte PRs in 30 T)

Beschreibung

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.

Contributor Guide