jmanhype/ane-lora-training

LoRA fine-tuning gradients on Apple Neural Engine via private APIs. Conv-as-matmul, spatial constraints, MLX integration.

6 best points · 1 comments · 1 Hacker News post · 10 GitHub stars · View on GitHub

Hacker News discussions

• Show HN: LoRA gradients on Apple's Neural Engine at 2.8W (6 points, 1 comments, by jmanhype, 2026-03-06)