DeepSeek V4.1-Flash-1017 Tops OpenRouter Weekly Usage Ranking
DeepSeek V4.1-Flash-1017 ranked first on OpenRouter’s Oct. 19–25 weekly model-usage chart with 33.6 trillion tokens, up 801% from the previous week. The previous V4.1 Flash ranked second with 18 trillion tokens and a 45% decline, while MiniMax M3.1 Flash, GLM 5.3 Flash, and Tencent Hy4 preview completed the top five. The ranking covers aggregated API traffic routed through OpenRouter and excludes vendor websites, independent apps, and private deployments.
The result is a material adoption signal for DeepSeek V4.1-Flash-1017 and shows how quickly usage can shift between model versions on OpenRouter, though it does not represent total global model traffic.
10月26日消息，在OpenRouter发布的最新一周全球AI大模型调用量周榜单上（10月19日至25日），拿下榜首的是DeepSeek在10月17日更新的V4.1-Flash-1017版本。该模型单周调用量达到33.6万亿Token，环比增幅高达801%，成为近一周爆发力最强的模型，也是拉动国产模型整体增量的核心动力。
公开资料显示，V4.1-Flash-1017以9月10日发布的V4.1 Flash为基底，延续552B总参数、输入激活8B、输出激活16B的非对称MoE架构，原生支持多模态，重点优化代码、自动化Agent与视觉任务。
OpenRouter单周调用量排名
榜单第二名是DeepSeek旧版V4.1 Flash，9月10日正式上线，本周周调用量18万亿Token，环比下滑45%，表明开发者群体快速向新版迁移。
#DeepSeek#DeepSeek V4.1-Flash-1017#DeepSeek V4.1 Flash#OpenRouter#MiniMax M3.1 Flash#GLM 5.3 Flash#Tencent Hy4 preview