gemini-4IMAGE
Gemini 4 Argon launches on ZenMux, claims top engineering reasoning score
ZenMux has listed Gemini 4 Argon for long-horizon coding and multi-step workflows, citing 77.9% on DeepSWE v1.1 and first place on the Vals Index at 68.90%.
xiax|AI & Agents
@xiax0603Archived 20 days agoOriginally posted 2026.10.13

xiax|AI & Agents @xiax0603Originally posted 2026.10.13
🚨突发!Gemini 4 Argon现已上线ZenMux,工程推理基准拿下榜单首位! Google Gemini 4 Argon已经在ZenMux正式上线,模型定位聚焦长时程编码、法律金融调研、网络防御多步骤复杂工作流。 核心公开成绩:DeepSWE v1.1真实软件工程基准77.9%,Vals Index综合排行榜#1,得分68.90%,适配仓库级调试、跨文件代码重构、漏洞发现修补。 开发者直接复用ZenMux统一API即可调用,无需改动现有接口。 完整链接: https://zenmux.ai/google/gemini-4-argon
x.com/xiax0603/status/2110019194167214410Archived 2026.10.14 00:21