ml-explore/mlx

[Feature Request] Fallback for custom kernels

已關閉

#3,240 建立於 2026年3月11日

 (4 則留言) (0 個反應) (0 位負責人)C++ (1,794 個分叉)batch import
enhancementgood first issue

倉庫指標

星標
 (26,216 顆星)
PR 合併指標
 (平均合併 6天 17小時) (30 天內合併 75 個 PR)

描述

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.

貢獻者指南