Performance

Performance

Headline numbers for v0.10.0. Two HIP ISAs share one tree; build with an explicit HIP_ARCH (Getting started). Wall time moves with page-cache state; treat E2E as a band. Denoise GPU (HIP events) is the stable GPU figure.

In-tree copy: docs/PERFORMANCE.md.

v0.10.0 (2026-09-01)

Preset	gfx1151	gfx90a
fox-s2 E2E	83–87 s	~10.5 s
fox-fast E2E	~95 s	~18 s
15 s cinematic E2E	45.0 min	12 min 33 s

gfx1151 fox-s2 md5 is unchanged from v0.9.0: 1731f95c4aa582597cf83d57f46b8f9e.

# fox-s2 — HIP A/B smoke (gfx1151 md5 gate)
./h3 --profile -d /path/to/MiniMax-H3 \
  -p "A red fox walks through fresh snow." \
  --width 512 --height 512 --frames 22 --steps 2 --layers 35 --reuse 1 \
  -o outputs/fox-s2.mp4

# fox-fast — upstream tutorial “first fast video”, without --show
./h3 --profile -d /path/to/MiniMax-H3 \
  -p "A red fox walks through fresh snow in a pine forest. Medium tracking shot, natural winter light, realistic fur, soft footsteps and wind." \
  --width 512 --height 512 --frames 22 --steps 20 --layers 45 --reuse 2 \
  -o outputs/fox-fast.mp4

--reuse 2 on 20 steps yields 11 DiT evaluations. Showcase fox uses --layers 50 --reuse 1 and is not the fox-fast scoreboard.

Long T2VA: Long video. gfx90a DiT is BF16 by default (H3_INT8_MLP=1 restores INT8).

gfx1151 history (v0.9.0)

When	fox-s2 E2E	fox-fast E2E	What landed
2026-08-18	~257 s	—	First HIP profile
2026-08-19	~117 s	—	INT8 DiT, faster load I/O
2026-08-22	—	~213 s	First fox-fast measure
v0.9.0	83–87 s	95 s	WMMA kernels; weights in VRAM carveout
v0.10.0	same on gfx1151	~18 s on gfx90a	Dual ISA; MI210 MFMA flash