MiniMax M3 带来稀疏注意力、1M 上下文和多模态能力,Together 承担了让它跑得更快的服务端工作。作者提到这是与 Together 团队的一次合作。
x.com/i/article/206189124776…
推荐理由:M3 同时列出稀疏注意力、1M 上下文与多模态三项能力,并点出 Together 承担了让它跑得更快的服务端工作。
@skylermiao7 · X · 历史来源 · 当前未持续收录
MiniMax M3 带来稀疏注意力、1M 上下文和多模态能力,Together 承担了让它跑得更快的服务端工作。作者提到这是与 Together 团队的一次合作。
x.com/i/article/206189124776…
推荐理由:M3 同时列出稀疏注意力、1M 上下文与多模态三项能力,并点出 Together 承担了让它跑得更快的服务端工作。
Introducing MiniMax M3: The First Open-Weights Model to Combine Three Frontier Capabilities - Coding & Agentic Frontier: 59.0% SWE-Bench Pro, 66.0% Terminal Bench 2.1, 34.8% SWE-fficiency, 28.8% KernelBench Hard, 74.2% MCP Atlas - MiniMax Sparse Attention scales context to 1M - Natively Multimodal from Step Zero API: platform.minimax.io Token Plan: platform.minimax.io/subscrib… 🚀New! MiniMax Code: code.minimax.io Weights & Tech Report in ~10 Days
Upgrading Agent — here's what's new 🧵 🤝 Agent Teams Multi-agent, multi-stage parallel execution on desktop. Adversarial quality gates between phases. Spin up agents across roles — run your own one-person company. 🧠 Mavis — your personal AI chief of staff Lark, WeChat, Telegram. Remote desktop control from your phone. Persistent memory that compounds the longer you use it. 🎯 One subscription, everything unlocked TokenPlan + Agent Plan merged. CLI, API, Agent — all in. Every model: M2.7, music, video, voice. Flexible credits shared across Agent and API. Already on both plans? Free extra month on us. 🔓 Open-source incoming — Teams, Mavis, all of it.