Ant Ling· @AntLingAGI · X·· 2 小时前精选AI 评分65
AI 导读
蚂蚁百灵转发 SGLang 对 Ling-3.1-flash 的 Day-0 支持。该模型为 560B MoE 混合推理模型,激活 25B,支持 1M 上下文;MXFP4 + block-FP8 量化将权重从约 534GB 降至约 292GB,可放入 4 张 Hopper GPU,并在 GDPVal-AA、FrontierSWE、Terminal-Bench 4.0 等真实 Agent、办公和编码任务上表现较强。
推荐理由
引用内容给出 Ling-3.1-flash 的参数量、量化方案和部署门槛,读者可据此评估在 4 张 Hopper GPU 上自部署的可行性。
正文 · 原文
Thanks @sgl_project for day-zero support and for helping developers serve Ling-3.1-flash more efficiently! ⚡
SGLang has Day-0 support for Ling-3.1-flash from @AntLingAGI! A 560B MoE hybrid reasoning model, 25B active, 1M context, built for real work: it turns a rough brief into a page, mobile UI, or interactive prototype. ✅ MXFP4 + block-FP8 quantization cuts weights from ~534GB to ~292GB, fitting on 4 Hopper GPUs ✅ Strong on real-world agent, office, and coding tasks, including GDPVal-AA, FrontierSWE, and Terminal-Bench 4.0. Get started with SGLang:在 X 查看被引用的帖子
来源:Ant Ling · x.com