Nace AI 发布多模态决策模型 Drex Max，可从截图直接做决策
Nace AI 出了个 Drex Max，看截图就能做决策，34 毫秒一次，Mind2Web 跑到 88.3%，做浏览器智能体的可以试试。
Nace AI 推出多模态决策模型 Drex Max，支持文本、图片和视频输入。模型可从截图、录屏、扫描文档、科学图表和照片直接生成决策，单次决策耗时 34 毫秒。在 Mind2Web 基准上得 88.3%，超过 Jev 1.14；Decision Index 得 67.6（Jev 1.14 为 60.70）。定价为每百万输入 token 0.06 美元，适配 computer-use 和 browser-use 智能体场景。
Decision models are going multimodal. Drex Max looks really strong. It can make fast decisions straight from screenshots, screencasts, scanned documents, scientific charts, and photos. This model is a really good fit for computer-use and browser-use agents, where every step starts with reading the screen. Nace reports 88.3% on Mind2Web (outperforms Jev 1.14). Drex Max also scores 67.6 on the Decision Index (Jev 1.14: 60.70). Your browser does not support the video tag. 🔗 View on Twitter Nace AI @NaceAI Drex Max is here: our multimodal decision model text, images, and video 34 ms/decision, under 1s at 262K tokens $0.06 / 1M input tokens available today at nace.ai/drex Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 4 🔄 0 ❤️ 16 👀 1952 📊 5 ⚡
- TestingCatalog14:17原文
