qwen2.5:14b on A100, H100 — 2026-04-18
2026-04-18T06:47:28.000Z → 2026-04-18T06:48:03.692Z
On 2026-04-18, qwen2.5:14b ran across A100, H100. H100 finished first at 105.9 tok/s, 14× faster than A100. The per-million-token cost: $6.53 (H100) vs $28.14 (A100) — the surprise: the headline GPU H100 was also the cheapest per-million tokens, coming in 4.3× less than A100.
Podium
H100
1st
thunder-h100
- peak tok/s
- 105.9
- avg tok/s
- 105.9
- $ / 1M tok
- $6.53
A100
2nd
thunder-a100
- peak tok/s
- 7.7
- avg tok/s
- 7.7
- $ / 1M tok
- $28.14