Primary sourceAI Will
View sourceEvent ·Multi-source
Kimi K3 Max在内部知识工作基准上超越Claude Opus和GPT-5.5
Kimi K3 Max在Kimi内部知识工作基准上获得三项得分：Online Exp Bench 75.5分，DECK-Bench 73.5分，Finance-Bench 62.6分。这些基准测试模拟真实用户Agent工作流中的常见任务。三项成绩均超过Claude Opus 4.8（max）和GPT-5.5（xhigh）。这表明Kimi K3 Max在Agentic知识工作上的能力和可靠性有所提升。
Current read
Kimi K3 Max在三个内部知识工作基准上全面超过Claude Opus和GPT-5.5，适合关注智能体工作流的人看看。
3 sources62° AI heatLast updated 2026/7/17 04:00:34
Evidence
3 sourcesConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.