@kimmonismus· @kimmonismus · X·· 2026-06-01精选AI 评分74
AI 导读
MiniMax 发布 M3 开放权重模型，在 SWE-Bench Pro 上得分 59%，略高于 GPT-5.5 的 58.6%，高于 Gemini 3.1 Pro 的 54.2%，编码上仍落后 Opus 4.7。该模型支持 1M token 上下文并原生多模态，在 BrowseComp 上以 83.5% 领先 Opus 4.7，每 token 成本约为 GPT-5.5 的 1/12。权重和完整技术报告预计约 10 天后发布。
推荐理由
原文给出了 M3 与 GPT-5.5、Opus 4.7 的基准对比和成本差距，读者可据此判断开放权重模型的能力位次。
正文
MiniMax just dropped M3! It hits 59% on SWE-Bench Pro, edging out GPT-5.5 (58.6%) and beating Gemini 3.1 Pro (54.2%).
Trails Opus 4.7 on coding, but leads it on autonomous browsing at 83.5% on BrowseComp. First open model to pack frontier coding, a 1M-token context, and native multimodality into one system.
I mean, let that sink in: Roughly 12x cheaper per token than GPT-5.5, with weights and a full tech report promised in about 10 days.
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在 X 查看被引用的帖子
来源：@kimmonismus · x.com