大模型
x.com ·
@googleaidevs
Google 演示用 Gemini 3.8 Live 构建屏幕感知编程辅导工具
Google AI Developers 展示了一个编程辅导示例：使用 Gemini 3.8 Live Extended Thinking 识别屏幕上的 bug，并通过 function calling 引用 p5.js 库，逐步讲解逻辑。该内容为官方演示视频，未提供实现细节或评测数据。
AI推荐理由 Google AI Developers 官方账号展示用 Gemini 3.8 Live Extended Thinking 构建可看屏幕、通过 function calling 调用 p5.js 库的编程辅导工具，属于官方演示，涉及多模态与函数调用能力。但正文仅为演示描述，无技术细节、代码或可复现证据，效果无法验证。
60
浏览 2,197
点赞 8
评论 2
转发 1
正文
Ever get stuck on a bug and wish the exact problem could be pointed out to you on screen?
Watch how we used Gemini 3.8 Live Extended Thinking to build a coding tutor that sees your screen and uses function calling to reference the p5.js library, and talks you through the logic.
[Your browser does not support the video tag.](https://video.twimg.com/amplify_video/2102106943984926720/vid/avc1/1920x1080/3sZU7qCGZKyT0dJq.mp4?tag=29)
[🔗 View on Twitter](https://x.com/googleaidevs/status/2102107233912016926/video/1)