Skip to content
llm-speed

H100 SXM vs MI300X

Side-by-side decode tok/s, prefill tok/s, and TTFT for H100 SXM and MI300X, sourced from community-submitted runs of the llm-speed suite. Every number on this page links back to the run it came from.

hardware
H100 SXM
NVIDIA Hopper · 80 GB VRAM
View H100 SXM page →
hardware
MI300X
AMD CDNA3 · 192 GB VRAM
View MI300X page →

No overlapping H100 SXM ↔ MI300X benchmarks yet.

We don't have a submitted run that covers both sides of this comparison yet. Run the suite on either side to populate this page:

$ pipx install llm-speed && llm-speed bench

See also: H100 SXM benchmarks · MI300X benchmarks · All hardware · All models · Methodology