Hand-written HIP kernel brings SageAttention to RX 9070 XT with nearly 2x speedup

A developer released a hand-written HIP kernel-based SageAttention build for AMD's RX 9070/9070 XT, delivering nearly 2x speedups in common ComfyUI scenarios, along with full build instructions and caveats for RDNA4 users.

2026-10-02 ~ 2026-10-02 · 2 related posts

• Getting SageAttention actually fast on RDNA4: a 9070 XT build guide with all the caveats — Familiar_Worry332 · 2026-10-02
• SageAttention up to 1.96x faster causal attention on RX 9070 XT via hand-written HIP fp8 kernel — Familiar_Worry332 · 2026-10-02