Google's Antigravity 2.0 introduces Gemini 3.5 Flash, a fast and affordable AI coding model that balances speed with quality output, along with new features including sub-agents for parallel task execution, scheduled task automation for recurring workflows, and a separated IDE model; however, the platform has limitations such as the removal of browser access due to privacy concerns and a less polished interface compared to competitors like Codex.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
Google Antigravity 2.0 Is Here: Gemini 3.5 Flash, Sub-Agents & Scheduled TasksAdded:
Google just released anti-gravity 2.0 and Gemini 3.5 Flash, which is super fast and very cheap to use. So, in today's video, I'm going to explain to you all you need to know about the new updates that Google just made. If you haven't subscribed to my YouTube channel, please remember to do so.
Recently, Google just released anti-gravity 2.0. And one of the key things that they did is that they separated the IDE model from the app. So basically now you have an IDE model where you can run different models like for instance you can run cloud code inside the IDE model. Some users can still access other models like cloud and GPT inside anti-gravity 2.0 new app but other users complain that they cannot use it. Another important thing that I've noticed with the Google anti-gravity 2.0 new model is that they have the Gemini 3.5 flash which is super fast. Honestly, it is one of the super fast models. I can compare it to Haiku, but I think it has a better output when you compare to Haiku, but it has a better speed when you compare to other models like GPT 5.5 or even Opas 4.6 or Opas 4.7. It is super fast and the output is nice. So, it kind of has like a balance in between quality work and also speed. Like for instance, I had it build a nice website for my car channel.
So this was the website that I was trying to build. I was trying to showcase some of the car videos that I have in my gallery. So it built a nice website. Honestly, it did this task within a few minutes and it was up and ready. Another thing that I've noticed with the new anti-gravity 2.0 is that now you can have sub agents running inside anti-gravity 2.0. So you can have different conversations or you can have multiple agents working on your task.
For instance, when I was trying to build this website, I deployed two sub agents that were building the website. So these sub agents that they added are very important because now they increase the speed at which it can accomplish tasks for you. Another thing that I noticed inside the anti-gravity 2.0 is that they have the schedule task right here. For instance, I have a task that I created today. It's a daily AI news update. So it basically runs at 5:00 a.m. and then obtains information related to AI. This is a nice update that can be utilized by business companies. Like for instance, if you have a company, you can set up automations that you can run at scheduled intervals in your organization. Uh I feel like they're trying to keep up with other models like Codex. When you look at this app, the anti-gravity 2.0 And then when you compare it to a Codex app right here, the interface is kind of the same, but Codex is more upgraded. I can say or I can say that Codex is much better because the user interface for Codex is much better. Like for instance, Codex has the plugins right here. These plugins are very important thing that they need to also add inside. I know it's new, but they need to include the plugins. Plugins are very important when you're building automations or agents because you need to connect to other external things. So earlier on when you were using anti-gravity IDE you had the option of using the browser like for instance if you were running a task using a Gemini 3.5 flash you were able to open the browser directly but now with this new app that they have just introduced you cannot this is something that is very limiting I think next month they're going to be releasing a Gemini 3.5 flash pro so hoping that when they do that they're going to take into consideration the need to have a browser use incorporated inside this app.
According to Google, the reason as to why they removed the browser use is because of privacy concerns. Now you probably have to install other tools like the ChromeDev tools, but I hope they're going to take into consideration. My honest key takeaways from the new updates is that Gemini 3.5 Flash is super fast, reliable, and also cheap. So if you are trying to get a balance between reliability, quality work and fast responses that is the best model to use. But we cannot compare to other models like for instance GPT 5.5 or Opus 4.7. These are specialized a models that Google still needs to keep up with. It is also worth noting that Google also included the schedule tasks that can allow you to set up automations that run on a regular basis. So that is it for today. Hopefully you found this video to be educative and also important. And if you happen to like it, please remember to like and subscribe to my YouTube channel.
Related Videos
OpenHuman VS Hermes AI: Who Wins?
JulianGoldieSEO
285 views•2026-05-29
Long-Running Agents — Build an Agent That Never Forgets with Google ADK
suryakunju
142 views•2026-05-30
This computer is made from real human brain cells. And you can buy it.
Talktmsmedia
3K views•2026-05-28
BREAKING: Microsoft’s New Image Generating Model Beat Out GPT 1.5 and Nano Banana 2
aimmediahouse
122 views•2026-06-03
I Made the Same Anime Fight Scene in Every AI Video Generator
NobleGooseAnime
295 views•2026-05-30
Nvidia Bets Big On AI PCs | New Chip To Power Windows Laptops | Technology | AI Updates | N18S
cnnnews18
3K views•2026-06-01
I Tested NEW Opus 4.8 on Four Projects (Updated LLM Leaderboard)
AICodingDaily
298 views•2026-05-29
3D Platformer Update - NO CAPES
SolarLune
294 views•2026-05-30











