蚂蚁百灵 Ling-3.1-flash 获 SGLang 首日支持
Thanks @sgl_project for day-zero support and for helping developers serve Ling-3.1-flash more efficiently! ⚡
蚂蚁百灵宣布 SGLang 在发布当天即支持 Ling-3.1-flash，帮助开发者更高效地部署该模型。据 SGLang 介绍，Ling-3.1-flash 是 560B MoE 混合推理模型，激活参数 25B，上下文窗口 1M，面向真实工作场景，可将粗略需求转化为网页、移动端 UI 或交互原型。其 MXFP4 加 block-FP8 量化把权重从约 534GB 降至约 292GB，可放入 4 张 Hopper GPU，并在 GDPVal-AA、FrontierSWE 和 Terminal-Bench 4.0 等真实智能体、办公与编码任务上表现较强。
查看原帖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…
展开完整原文
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:
正文
当前显示原文，中文全文译文尚未就绪。
Thanks @sgl_project for day-zero support and for helping developers serve Ling-3.1-flash more efficiently! ⚡