Sparse-Attention on Scholionhttps://scholion.thluiz.com/tags/sparse-attention/Recent content in Sparse-Attention on ScholionHugo -- gohugo.iopt-BR© 2026Fri, 25 Sep 2026 11:42:27 +0100MiniMax M3 and the Cost of Long-Context Agentshttps://scholion.thluiz.com/notes/minimax-m3-cost-of-long-context-agents/Wed, 23 Sep 2026 13:07:06 +0100https://scholion.thluiz.com/notes/minimax-m3-cost-of-long-context-agents/MiniMax M3’s sparse attention keeps per-step cost flat as context grows, making multi-hour autonomous agents affordable in production.