PodcastIntel
Sign in Get Started Free
AI Engineer
AI Engineer

How MiniMax M3 Was Built: Sparse Attention and Native Multimodality — Olive Song

Oct 10, 2026 · 0:15:03
✨ AI Summary
  • MiniMax M3 natively handles text and vision from training start.
  • Sparse attention (MSA) offers significant speed improvements over full attention.
  • Achieve frontier-level coding and agentic abilities with a 1M token context.

Guests on This Episode

OS
Olive Song
2 podcast appearances

More from AI Engineer

View all episodes →

Get AI Summaries for Every New Episode

Subscribe to AI Engineer and get AI summaries, guest tracking, and email digests delivered automatically.

Sign Up Free →