Vera Rubin inference per watt vs Blackwell — up to 10× better
up to 10×estimate
| Value kind | estimate — Interpret this value according to its displayed kind, scope, source, and as-of date. |
|---|---|
| Caveat | 'Up to' vendor figure; the release names no baseline (context implies Blackwell). The product page states the comparable figure as tokens per megawatt vs GB200 NVL72 on Kimi-K2-Thinking 32K/8K. Ratio only. |
| As of | 2026-03 |
| Source | NVIDIA newsroom, Vera Rubin platform release (GTC, 2026-03-16) · Release body: Vera Rubin NVL72 'up to 10x higher inference throughput per watt' at one-tenth the cost per token |
| Review | checking…review by 2027-01-17 · fast cadence |
| Recorded changes | last 2026-10-19 · 2 revisions tracked |
| Claim id | vera-rubin-inference-per-watt-vs-blackwell |
Where the guide uses it
Not quoted in a chapter yet; it is kept in the curated register.
← Full numbers register — every date-stamped figure in the guide, with revision history.