LingBot-Video 获 SGLang v0.5.18 原生支持
Huge shout‑out to the @sgl_project community! 🙌 LingBot‑Video gets native support in SGLang v0.5.18.
非常感谢 @sgl_project 社区！🙌 LingBot-Video 在 SGLang v0.5.18 中获得原生支持。用优化后的 SGLang-Diffusion 服务运行我们的 30B-A3B MoE 文生视频模型。Cookbook👉 https://t.co/oz2ssi9Chs
查看原帖SGLang v0.5.18 is out! We have 710 PRs from 212 contributors, and 61 of them were new contributors. Welcome🧡 Highlights: 🔸 New models: Muse Glimmer and Intern-S2-Mobius, plus SANA-Video, LTX-2.5, LingBot-Video, Cosmos3, and LongCat-Image on SGLang-Diffusion 🔸 Engine startup is up to 2.38x faster: checkpoints stream from disk while CUDA graphs capture,…
展开完整原文
SGLang v0.5.18 is out! We have 710 PRs from 212 contributors, and 61 of them were new contributors. Welcome🧡 Highlights: 🔸 New models: Muse Glimmer and Intern-S2-Mobius, plus SANA-Video, LTX-2.5, LingBot-Video, Cosmos3, and LongCat-Image on SGLang-Diffusion 🔸 Engine startup is up to 2.38x faster: checkpoints stream from disk while CUDA graphs capture, so the two longest waits happen at once 🔸 All compiled kernels (Triton, FlashInfer, Inductor, DeepGEMM) now live in one directory: SGLANG_CACHE_DIR 🔸 NVFP4 checkpoints now run on AMD GPUs via online MXFP4 requantization, with 97.5-100% accuracy retention across MiniMax-M2.7, GLM-5.1, Kimi-K2.6, Qwen3.5-397B, and DeepSeek-R1 🔸 Kimi K3 runs much faster on AMD MI355X: 1.37-1.77x throughput and up to 2.42x faster ITL 🔸 DeepSeek-V4 decode gets faster on NVIDIA: TP LMHead all-to-all cuts LMHead time from 320us to 169us, and FlashInfer MNNVL allreduce turns on automatically for DeepSeek models Thanks again to our amazing partners and model makers: @NVIDIAAI @AIatAMD @intel @AIatMeta @intern_lm @Lightricks @AntLingAGI @robbyant_brain @Meituan_LongCat Full release note👇
正文
当前显示原文，中文全文译文尚未就绪。
Huge shout‑out to the @sgl_project community! 🙌 LingBot‑Video gets native support in SGLang v0.5.18. Run our 30B‑A3B MoE text‑to‑video model with optimized SGLang‑Diffusion serving. Cookbook👉 https://t.co/oz2ssi9Chs