← back to paper

Review history

arxiv: 2605.09825 · 3 revisions

Pretraining large language models with MXFP4 on Native FP4 Hardware

1. 2026-05-15 UNVERDICTED LOW v0.9.0 novelty 6.0
   57296 ms 5526 in 1061 out 2026-05-15T05:12:36.108870+00:00
2. 2026-05-14 UNVERDICTED LOW v0.9.0 novelty 6.0
   49655 ms 5526 in 969 out 2026-05-14T21:01:21.753046+00:00
3. 2026-05-12 UNVERDICTED LOW v0.9.0 novelty 7.0
   69754 ms 5521 in 1106 out 2026-05-12T04:58:16.324360+00:00