DCGM level-4 planning runtime estimates for distinct four- and eight-GPU Hopper configurations
under 45 min (4 GPU) / under 2.25 hr (8 GPU)estimate
| Value kind | estimate — Interpret this value according to its displayed kind, scope, source, and as-of date. |
|---|---|
| Scope | Separate four- and eight-GPU Hopper estimates, not a pooled range or fixed per-node runtime. Qualify the installed release, hardware, selected tests and exclusivity; a level-4 result does not establish EUD execution. The same table gives level 2 under 2.5/10.5 min and level 3 under 10/35 min for four/eight GPUs respectively; these are separate planning estimates. |
| As of | 2026-09 |
| Source | NVIDIA, DCGM Diagnostics, run-level table (rolling documentation). · Hopper-system run-level runtime table |
| Review | checking…review by 2027-03-08 · standard cadence |
| Recorded changes | last 2026-09-16 · 3 revisions tracked |
| Claim id | dcgm-r-4-deep-incl-memtest-eud-runtime-per-node |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.