Tabbit

Meet Tabbit—the Agentic AI browser built for you. Now available on macOS and Windows.

@adelbucetta We agree this is an important question. We’re working to give users more control over their workflows and make it clearer what’s happening at each step. We’d love to hear your feedback as you use Tabbit.

@TabbitBrowser tabbit v1.4 looks like a good step towards making ai tools more manageable, but what's still unclear is how users will navigate tradeoffs between convenience and data quality in these integrated workflows

What's New in Tabbit V1.4 Files is now on the homepage, giving you one place to manage Tabbit-generated files. Deep Research is now available from Task Shortcuts during chat. Powered by the latest GPT-5.6. Try it now.

@gegenpressboss Agent Mode uses our top-performing model, which we’ll update regularly.

@TabbitBrowser Does the Agent Mode use the selected model or what?

GPT-5.6 Sol, Terra, and Luna are now on Tabbit. Try them for free during our public bonus period.

@AbionMorse Set Tabbit as your default browser to claim pro membership for free during our public bonus period!

@TabbitBrowser how much is the pro membership and what does it include? (link?)

A wrap-up of our latest models. Free to use. Claude Opus 4.8 GPT 5.5 GLM-5.2 Gemini-3.1 Pro Qwen3.7-Max And more. Set Tabbit as your default browser to claim free Pro membership and use all models at no cost during our public bonus period.

Sorry for the bad experience, and thanks for letting us know. That’s not the experience we want anyone to have. Tabbit AI chat is currently only available inside the Tabbit Browser app, not on a website. Could you share where you saw that chat-like interface? We’d really like to look into it.

@TabbitBrowser I came across your browser through a website - ai chat like interface when ever i tried to enter message and submit that it was forcing me to install your browser. That was very bad experience. I immediately uninstalled because your marketing was just invasive and seemed scam

@AliceInfoAi 🧡

@TabbitBrowser Would love to try

Claude Sonnet 5 is now live in Tabbit. In Tabbit, it understands your current browser context, making research, coding, and everyday workflows faster and smarter. Available now for free. Tap “Switch Model” to try it.

@_Dettol99 The international mobile version of Tabbit is still in development. Stay tuned!

@TabbitBrowser I downloaded the iPhone application, but it doesn't work until China's number. Is there a solution?

@ampremzz67587 Copy that.

@TabbitBrowser Please make android and IOS version

@highsun1499 Set Tabbit as your default browser to claim Pro for free and use premium models during public bonus period.

@TabbitBrowser How long will the premium models be offered for free?

Meet Tabbit—the Agentic AI browser built for you. Now available on macOS and Windows.

@RaveyFok 感谢喜欢！您的反馈已收到，已同步产研团队

@TabbitBrowser 我非常喜欢 Tabbit 这个浏览器，你们上新模型的速度也很迅速，可是使用 tabbit 真的完成本地小项目的时候他总是卡在沙箱里出不来，通常他花费的时间很长很长最后会报错退出

@xieyibiao1 您好，在设置页面您每周可以免费领取重置券，即可再次获取一管 100% 用量

@TabbitBrowser 你好，我把这个设置为“默认浏览器”了，但是我的用量没有了，我要如何提升用量？

@LZRationalnvest Then it’s time to try Tabbit. Claude Sonnet 5 for free now.

我好奇的是，Claude封了那么多中国人的号，那些没被封号的，还在中国大陆使用的Claude用户，究竟是何方神圣？😇

Kimi K2.7 Code is now live in Tabbit. Kimi's smartest coding model yet. Inside Tabbit, it stays locked on context through every follow-up and debug, so code, docs, and dev problems get solved faster. Free to try now. Click Switch Model at the bottom right of the chat box.

@jimmyzhou2023 @fengdu2077 您好，目前 tabbit 移动端国际版还在开发中。您可以切换至 PC 端使用相应模型。

@fengdu2077 iOS的tabbit 为啥没有Claude 、Gemini、GPT? 打开方式不对吗

分享一个免费用 Claude 、Gemini、GPT的方法，都是可用他们的最新旗舰模型。 除了御三家，还有国内的明星模型：Kimi-2.6、GLM-5.1、MiniMax-M3 只要下载一个美团的「tabbit 国际版」，记住要用国际版的才有御三家。 用国内版的只有国内的模型。

@izzy2fx @AYi_AInotes Yes, users can manually select models based on the specific tasks they are running. You can also select multi-model mode to check the performance from different models.

@AYi_AInotes How does tabbit international achieve seamless switching among multiple flagship models? Can users manually select the most suitable model based on task type, or is it system-assigned? How do different models perform on complex reasoning and code generation tasks?

我去，GPT 5.5， Claude Opus 4.8， Gemini 3.5flash，最新旗舰版全接入，完全免费不用单独开订阅！ 美团最近悄悄上了一个叫 tabbit 国际版的应用，里面接了好几家顶级模型，Claude、Gemini、GPT，都是最新旗舰版。 国内这边也没落下，Kimi、GLM、MiniMax 都在。 唯一的坑：一定要下国际版， 国内版只有国内模型，御三家不接入， 这东西的逻辑很简单，美团想做 AI 入口，先砸钱把模型阵容堆满，把用户量拉起来， 现阶段一分钱不花就能同时用几家最贵的模型，之前不想在各家之间来回切，又不想开一堆订阅的人，这个方案挺省心的， 趁现在还没收费，先用着，谢谢兄弟@fengdu2077挖掘到这么好的工具！亲测有效！

@fakechuanpu2000 @AYi_AInotes 您好，设置成默认浏览器即可免费领取 Pro 会员，并使用相应模型

@AYi_AInotes 装了国际版也不行啊？

GLM-5.2 from @zai_org is now live in Tabbit. A 1M context window for long, complex tasks, and the No.1 open-source model on Code Arena. Paired with Tabbit's multi-turn conversations, it breaks big problems down and works through them step by step. Free to try now.