Compare
Side-by-side spec sheets for chips, systems and models. Every figure keeps its basis label — a vendor claim next to an estimate is labeled as exactly that.
| DeepSeek-V3 671B parameters, $5.6M disclosed final-run cost — the efficiency shock. | Llama 3.1 405B The first open-weights model at closed-frontier scale. | |
|---|---|---|
| Overview — Training | 14.8T tokens · 2,048 × H800 · ~2.79M GPU-hours (disclosed)Published paper | — |
| Overview — Disclosed cost | $5.576M final run only — excludes research, ablations, data and hardwarePublished paper | — |
| Overview — Training compute | — | ~3.8×10²⁵ FLOP (from disclosed GPU-hours)Published paper |
| Overview — Training hardware | — | ~16,000 × H100 (two 24k clusters built)Published paper |
"—" means that record does not publish this line; different vendors disclose different things, and we do not fill gaps with guesses.