跳到正文
原文
LMSYS:Blog(Chatbot Arena 团队)·· 12 小时前AI 评分51

SGLang 集成 DSpark:基于置信度的变长验证投机解码

Blog DSpark in SGLang: Speculative Decoding with Confidence-Driven, Variable-Length Verification Speculative decoding trades extra compute for fewer decode steps, and the trade sours as load grows: at batch size B with K speculative tokens the target verifies B K tokens every step, and past a poi... SGLang Team July 6, 2026

AI 导读

SGLang 宣布正式支持 DSpark 投机解码算法(PR sgl-project/sglang#30261),覆盖 Qwen3 等稠密模型和 DeepSeek-V4 等稀疏模型。

来源:LMSYS:Blog(Chatbot Arena 团队) · lmsys.org