Same target model, same prompts, same tree settings, same H200 doing the verify. The only difference: who builds the draft tree, four Alveo V80 cards or the H200 itself. Humaneval, 80 prompts × 256 new tokens, tree depth 15, width 2, 20 verify tokens, fa4 draft attention backend on both sides.
| Tree | Draft backend | FPGA tok/s | FPGA runs | FPGA accept | H200 tok/s | H200 runs | H200 accept | FPGA ÷ H200 |
|---|
--draft-tree-depth --draft-tree-width 2 --verify-tokens-num --speculative-draft-attention-backend; FPGA = --speculative-exec-mode 1, H200 alone = -1. The FPGA side also uses a card-side tree-stop threshold (−2.5), which has no GPU equivalent.F3T_CONFIG_SWEEP_20261109.md §1 and the 11-10 sweep (tracker line 1585). H200 alone: same report §2 and demo_cfg_20261110/gpu3.out (11-10 19:35–19:44Z). Values with several runs are means.