v0.5.22
中文摘要
v0.5.22 更新引入了 Clef、LLaDA2.2 等多个新 LLM/VLM 及扩散模型,并显著提升了 DeepSeek-V4.1 在 H200 和 AMD 上的吞吐量。
English Summary
v0.5.22 introduces new LLM/VLM and Diffusion models like Clef and LLaDA2.2, and boosts DeepSeek-V4.1 throughput on H200 and AMD GPUs.
Highlights 950 PRs from 245 contributors. New models Model Type Cookbook Clef / Clef Flash LLM / VLM link PPLX-Decider v1 / v1.1 LLM / VLM link EmbeddingGemma 2 LLM / VLM link LLaDA2.2 LLM / VLM HunyuanImage-3.0-Instruct Diffusion link Kandinsky 6 / Kandinsky 6 SR Diffusion link Wan-Animate-2 Diffusion link SANA-Video 2.0 Diffusion link LLaDA-Image / LLaDA-Image-Turbo Diffusion What's new DeepSeek-V4.1 gets up to 2.3x more throughput per GPU on H200, and now runs on AMD MI350X / MI355X with up to 1.9x throughput from DSpark (#41251, #41308). Qwen3.8-Flash-Next decodes 12% faster on B200, and GLM-5.3-Flash with DFlash 10% faster on B300 (#42175, #42489). MiMo-V2.6-Pro gets 1.85x decode throughput with DFlash on B300 (#41886). Models load 10x faster with the new InstantTensor loader (#40453). PD disaggregation gets up to 27% more throughput and half the p95 TTFT on long-prompt workloads (#40703). HiSparse now supports EAGLE and MTP for long-context serving with the KV cache in CPU memory (#37771). GLM-5.2 and GLM-5.3 can now combine decode context parallelism with MTP, with up to 2.1x faster decoding on GLM-5.3 (#43735). SGLang now runs on Apple Silicon Macs with the standard PyTorch…