BLAST RADIUS

Open-weight & owned stack · Quantization efficiency

Splash Demonstrates Chip-Specific Kernel Tuning Gains

Sep 25, 2026

Splash reportedly increased decode speed by 20% on an M5 Max by selecting chip-specific matrix-multiplication layouts.

Read the original at reddit.comOpens the publisher's site in a new tab

More in Open-weight & owned stack