Box 实测 Gemini 3.5 Flash-Lite 在企业文档任务上比上代强了 17 个点，速度快还更轻，做数据分析和金融场景的可以关注。
Google DeepMind 发布了 Gemini 3.6 Flash 和 Gemini 3.5 Flash-Lite。Box 公司对 3.5 Flash-Lite 与 3.1 Flash-Lite 进行企业文档任务评估，总体准确率从 41% 提升至 58%。数据分析任务提升 27 个百分点，零售任务提升 28 个百分点，金融服务任务提升 16 个百分点，消费品任务提升 21 个百分点。新模型在更轻量更快速的同时实现了性能代际提升。
https://t.co/ewfS2ai4uo
x.com/Box/status/207… Box @Box . @GoogleDeepMind released Gemini 3.6 Flash and Gemini 3.5 Flash-Lite. We ran evals. Here is how 3.5 Flash-Lite compared to 3.1 Flash-Lite on real enterprise document tasks: → Overall: 58% vs 41% → Data Analysis: +27pp → Retail: +28pp → Financial Services: +16pp → Consumer Products: +21pp A clear generational step up on the hard, quantitative work that drives real decisions — and it gets there faster and lighter than the model it replaces, not slower. 🔗 View Quoted Tweet 💬 1 🔄 0 ❤️ 0 👀 458 📊 1 ⚡