Engineering isn’t just about faster typing; it’s about better thinking. Our team is excited to showcase Gemini 3.1 Pro, our best model for systemic reasoning. Watch as the model handles a complex "Legacy Sync" migration, not by just writing functions, but by identifying architectural risks. That’s the difference between... a chatbot and an engineering partner.show more

HELP.....Why is it doing that for me?

Fix the damn pro plan, we are paying for your service ! That's unacceptable🤬

Can anyone explain WHAT FUCK IS THIS? 5 days ban in shitty models? I've been using 3 flash all daylong. I give up. Go fuck yourself @Google

Waiting on the update still :(

Fix the bugs in the IDE and ship updates faster, clearly the people that work on this don’t even use it

You need to fix the quota issue for pro users, at least display a weekly limit of some sorts.😡😡

Antigravity did not auto update, I had to redownload it. Hadn’t done 1/10 of what I usually do with Gemini 3 pro which used to refresh every 5 hrs. And now you’re telling me to wait for more than 2 full days??

The 5hr refresh for pro users is now existent. It’s ridiculously easy to exceed my quota now, and you’re saying I should come back in almost 3 days??? Gosh

This is the worst release so far after the first one. No Antigravity auto update. I can't manually update. I can't use older models. I can't use new models. The worst job.

Gemini 3.1 Pro's migration strategy is next-level! Spotting those architectural risks while refactoring legacy sync? That's true engineering partnership, not just code spitting. Massive props to the team for pushing systemic reasoning boundaries. Tried it on a similar refactor yesterday—blown away by the depth.

Antigravity team, since you are started banning people for using your oauth key in other IDEs, i have a question. I don't care about other IDEs. I'm using oauth key in my own version of Can i use it, that's not prohibited? I'm not using it with other IDE's i'm using in my own personal scripts. Its for multi research with multi agents in parallel work, very helpfull. It's better than one agent in hard sciencific task. (Since Gemini co-scientist is not avialable for regular users). If not - do something similar. Since Antigravity can use parallel agents so this is possible.

I tried it but felt it didn’t perform as well as you hyped it. Google AI Studio didn’t even start building. In Antigravity it made a lot of mistakes.

antigravity is buggy, fix your shell exec problem in agent

"Better thinking, not faster typing" is the right framing. The models that win aren't the ones that output the most code. They're the ones that understand the problem first.

I only opened my Antigravity and had this! As Pro user this is so frustrating @antigravity. Not even touched this tool today.

The app is fucking shite lads. Constant issues.

Too bad antigravity still limits access to models for paying customers.

Gemini 3.1 Pro really shows what systemic reasoning looks like in practice.

Almost perfect. If you can improve compliance with / project rules (especially precedence + edge cases), it would be spot on.

made some cool stuff with gemini 3.1 in antigravity and also in aistudio :)

Still alot to improve for you guys

Shitty upgrade, with even worse rate limits. You can shut down the project. It's not our problem if you lack the competence to give us a proper functional product.

This is seriously impressive—autonomous dependency graphing and conflict resolution at this level feels like a real leap toward AI as a true engineering partner.

sorry, but its garbage. Whatever shenanigans its doing with multiple agents is really not working well.

how tone deaf is Antigravity team

This on didn't work either with my usage, I am on my 6th family plan to even code "hello world" maybe I should hire a human.

I want to see 3.1 pro reasonates live in action more!

Code generation is easy. Understanding the blast radius of that code is hard.

"Sequences shorted" -- Until agentic coding works at that speed, I still feel the urge to just do it myself instead of powerlessly watching the endless "thinking" messages.

Can you first fix the frequent errors that happen when calling AI models during programming? This really hurts the user experience.

Hey Google Antigravity, pls listing to your users feedback about Gemini 3.1 pro

Got annoying bugs. Everytime i ask anything in the chat, the IDE freezez until the answer or whole coding process is finished

Please fix the "There was an unexpected issue setting up your account. Please try again later." issue? Can't login into my account

Migrating legacy sync is usually a nightmare of hidden dependencies. Can this model actually map the side effects or is it just guessing based on the function names?

The potential is huge!! I thought gemini is only good at frontend work But this honestly changes my mind If it can do complex tasks seamless, then that puts it way ahead in the game Props to you guys!! Great work 🙌