dgx spark prefill speed

1 article

Estimating TTFT From FLOPs: The Prefill Compute Floor

Derive a time-to-first-token floor from model FLOPs, dense GPU TFLOPS and MFU, with the attention term, the sparsity trap and H100 worked examples.

October 28, 2026• 15 min read
ttft llmttft itldgx spark prefill speed