inference/research
verifiedhigh · 0.92updated 2026-10-03

MiniMax-01: Scaling Foundation Models with Lightning Attention

paper

Fields

id
sources/arch-minimax-01
URL
https://arxiv.org/abs/2501.08313
Publisher
MiniMax (arXiv)
Kind
paper
Published
2025-01-14
Accessed
2026-10-03
notes
Hybrid lightning-attention + MoE at 456B total / 45.9B active, 32 experts.

Notes

Hybrid lightning-attention + MoE at 456B total / 45.9B active, 32 experts.

Cross-references

This record references no other records.