FLOPS · Floating-Point Operations Per Second
The standard measure of compute throughput; AI clusters are rated in petaFLOPS and exaFLOPS.
Also written as: FLOP · FLOP/s · FLOPs
Current numbers
37–66%MI300X realized inference performance vs H100/H200 despite ~1.3x paper FLOPS — the ROCm goodput gap
37–66%MI300X realized inference throughput vs H100/H200 (2025), despite ~1.3x paper FLOPS — the historical realized-MFU gap
~30–50%typical model-FLOPS-utilization (MFU) for large LLM training; best-in-class >50% on Hopper