total traffic cut for SparseLoCo (H=15, 2-bit) vs 8-bit DiLoCo and 16-bit AdamW DDP — derived from Table 1 message sizes × sync counts; density-dependent (3.12% → 0.78%)
~30–94x / ~890–2,810xobserved
| Value kind | observed — Reported measurements, counts, and specifications keep the precision and scope stated by their source; an exact specification is not treated as a range. |
|---|---|
| As of | 2025 |
| Source | SparseLoCo v3 (arXiv:2508.15706), Table 1 |
| Review | checking…review by 2027-02-19 · slow cadence |
| Recorded changes | last 2026-08-31 · 3 revisions tracked |
| Claim id | further-traffic-cut-vs-diloco-vs-standard-ddp |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.