StepFun· @StepFun_ai · X·· 2026-06-04精选AI 评分77
AI 导读
阶跃星辰的 Step 3.7 Flash 已上架 Fireworks AI。该模型为 198B 稀疏 MoE 多模态大模型（VLM），含 196B 语言骨干和 1.8B 视觉编码器，从设计之初优化推理效率，采用硬件友好架构与 MTP 辅助解码，速度达 400 tokens/s。具备原生多模态理解与行动、可靠工具使用、增强搜索能力，面向真实智能体工作负载，采用 Apache 2.0 开源许可。
推荐理由
198B稀疏MoE加MTP解码把速度推到400 tok/s，还开源Apache 2.0，这规格做agent的大脑正合适，做实时应用的可以试试手。
正文 · 原文
Great to see Step 3.7 Flash live on @FireworksAI_HQ.
Designed for inference from day one, Step 3.7 Flash combines a hardware-friendly architecture with MTP-assisted decoding to reach up to 400 tokens/s.
Fast, multimodal, and ready to power capable agents in real-world workflows.
Many research labs only consider inference efficiency after the fact. Step 3.7 Flash is a 198B sparse MoE VLM designed by @StepFun_ai for inference from the start. 196B language backbone with a 1.8B vision encoder. Built for real-world agent workloads, running at up to 400 tok/sec. Native multimodal understanding and action, reliable tool use, and enhanced web and visual search. Apache 2.0. Try it now → https://fireworks.ai/models/fireworks/step-3p7-flash-nvfp4在 X 查看被引用的帖子
来源：StepFun · x.com