Hands-on: Gemini 3.7 Flash is fast, but details lag
A Juejin blogger tested Google's newly updated Gemini 3.7 Flash inside Antigravity and reported that most tasks finished in under three minutes, compared with over an hour for GLM 5.3 he used as a baseline; however, the cyberpunk web page lacked detail, the astronomical watch render was missing hands, and the 3D pool game never got the table's six pockets right. The post cites a DeepMind announcement saying the model is stronger at coding, knowledge work and web development, and notes its benchmark screenshots for FrontierCode, DeepSWE, AutomationBench and Code Arena compare against second-tier rivals such as Sonnet 5 and Terra rather than top-tier models.
The blogger says Gemini 3.7 Flash does little planning, tool calling or self-verification and produces usable code in a single fast pass, a notably different approach from mainstream agent-oriented models.
美版豆包又更新了！
还有人记得谷歌的 Gemini 系列么？好像被极度边缘化了！上次 Flash 更新的时候，我其实专门写过一系列文章，介绍 Antigravity 这个智能体，以及测试了 3.5 Flash。当前的感觉是 Flash 的前端设计还是很屌的，然后我就再也没关注过了！
看到 3.7 Flash 更新了，我又打开了尘封已久的 Antigravity，又折腾了一下网络。用了一下发现一种很神奇的感觉。
先说一下我的第一感受，Flash 真的是快到飞起，好像完全不用思考，也不需要纠错，一个任务给它，它刷的一下做完了。
GLM 5.3 需要一小时完成一个任务，Gemini 3.7 Flash 三分钟搞定。我测试了好多个例子，大部分例子没有超过 3 分钟。唯一一个让它用了 4 分钟的是超级玛丽。
#Google#Google DeepMind#Gemini 3.7 Flash#Antigravity#GLM-5.3#Claude Sonnet 5#GPT-5.6 Terra