Just when the tech world thought Anthropic’s Claude Code and OpenAI’s Codex had locked down the AI-driven development space, Google went ahead and flipped the script.

While competitors have spent the last year refining incremental autocomplete and context windows, Antigravity 2.0 has quietly rewritten the rules of the game.

Here is exactly how Google’s latest powerhouse is beating the industry’s favorite dev tools at their own game.

What is Antigravity 2.0, anyway?

Let’s understand the context here

When Google first dropped Antigravity last year, I will admit I was intrigued but thoroughly confused. While it’s basically a VS Code fork, Antigravity had this dual identity: a standard text editor and a futuristic Agent Manager, which looks like a totally different app.

Trying to split my brain between writing code manually and monitoring background AI agents in the same cluttered window felt like a massive challenge.

Google understood the identity crisis, and at I/O 2026, the company announced version 2.0. Now, the Antigravity software is split into two parts: a dedicated Agent Manager that looks and feels a lot like Claude Code and Codex and an Antigravity IDE tool which feels like VS Code with the Gemini add-on.

The best part of this pivot is the new UI. It is streamlined, responsive, and well-built. More importantly for my daily workflow, it is lightweight too. Unlike Codex, which regularly turns my laptop into a space heater and sets the cooling fans to maximum blast, Antigravity 2.0 runs completely cool in the background, even when it’s dealing with complex, parallel tasks.

Antigravity 2.0 features

Don’t forget about speed, too

Using the new standalone interface is a night-and-day difference from the cluttered, split-screen identity crisis of the old app. Antigravity 2.0 acts like a high-level command center. Google has packed a ton of agentic power into this release, and a few specific features redefine the workflow for me.

The biggest benefit is how it handles my files. Version 2.0 introduces an abstracted UI built around Projects. Now, a single project workspace can span multiple, totally separate folders across my machine.

It gives me the freedom to let an agent look at a frontend repo, a backend service, and a documentation folder all at once.

With the old app, if I gave the AI a massive, multi-part task, it would try to solve everything in a single chat. The context window would get bloated with terminal logs, and the model would eventually lose the plot.

Antigravity 2.0 solves this with dynamic subagents. The main orchestrator looks at a complex problem, breaks it down, and spins up subagents to tackle different parts in parallel.

👁 XDA
Quiz
8 Questions · Test Your Knowledge

How much do you know about Claude?
Trivia challenge

Think you know Anthropic's AI assistant? Put your knowledge of Claude to the test.

OriginsCapabilitiesSafetyFeaturesDesign
01 / 8Origins

Which company created Claude?

Correct! Claude was created by Anthropic, an AI safety company founded in 2021. Anthropic was co-founded by Dario Amodei and Daniela Amodei, among others who previously worked at OpenAI.
Not quite. Claude is made by Anthropic, not to be confused with OpenAI, which makes ChatGPT. Anthropic was founded in 2021 with a strong focus on AI safety research.
02 / 8Safety

What is the name of the safety and values framework Anthropic developed to guide Claude's behavior?

Correct! Anthropic developed Constitutional AI (CAI), a technique that trains Claude using a set of principles — a 'constitution' — to guide its responses toward being helpful, harmless, and honest.
Not quite. The framework is called Constitutional AI (CAI). It is a novel training approach pioneered by Anthropic that uses a written set of principles to help the model self-critique and improve its own outputs.
03 / 8Origins

What is the name most commonly associated with inspiring Claude's name?

Correct! Claude Shannon is widely cited as the inspiration behind the name. Shannon founded information theory, which is foundational to all modern computing and digital communication — a fitting namesake for an AI.
Not quite. The name Claude is most commonly associated with Claude Shannon, the mathematician and electrical engineer who founded information theory. His pioneering work laid the groundwork for the digital age.
04 / 8Capabilities

Which of the following best describes Claude's context window capability in its more advanced versions?

Correct! Advanced versions of Claude support context windows of 100,000 tokens or more, allowing it to process entire books, lengthy codebases, or large documents in a single conversation — a standout feature at the time of its release.
Not quite. Claude's advanced versions support context windows of 100,000 tokens or more. This was a significant leap beyond many contemporaries and allows Claude to reason over very large amounts of text in one session.
05 / 8Design

Which of the following principles is NOT part of Anthropic's core goal for Claude?

Correct! Anthropic's guiding principles for Claude are to be Helpful, Harmless, and Honest — often called the 'three H's.' Hierarchical is not part of this framework. The goal is to make AI that is safe and beneficial for everyone.
Not quite. Anthropic's three guiding principles for Claude are Helpful, Harmless, and Honest. 'Hierarchical' is not one of them. These three H's shape how Claude is trained to interact with users responsibly.
06 / 8Features

What was a key distinguishing feature of Claude 2 when it launched compared to many rival models at the time?

Correct! Claude 2 launched with a 100,000-token context window, which was remarkable at the time. This allowed users to feed in entire books or massive codebases for analysis, setting Claude apart from many competing models.
Not quite. The standout feature of Claude 2 was its 100,000-token context window. Claude does not natively generate images, and real-time browsing and built-in voice were not launch features of Claude 2.
07 / 8Safety

Anthropic describes itself primarily as which type of company?

Correct! Anthropic describes itself as an AI safety and research company. Unlike some competitors who lead with products or platforms, Anthropic's founding mission centers on building AI systems that are safe, interpretable, and steerable.
Not quite. Anthropic is primarily an AI safety and research company. Its founding mission is rooted in making AI that is safe and understandable, which is why safety-focused training methods like Constitutional AI are central to its work.
08 / 8Features

Which of the following tasks is Claude specifically designed to handle well?

Correct! Claude excels at long-form writing, summarization, coding assistance, and complex reasoning tasks. Its large context window and nuanced language understanding make it particularly well suited for handling detailed, multi-step text-based work.
Not quite. Claude is designed for text-based tasks like writing, summarization, analysis, and reasoning. It does not render graphics, autonomously execute system commands, or perform live video analysis — it is a large language model at its core.
Challenge Complete

Your Score

/ 8

Thanks for playing!

Scheduled Tasks is where the futuristic workflow really clicks. I no longer have to sit down, prompt the AI, and wait for a response. I can use Scheduled Tasks to automate routine background checks. For example, I can configure an agent to wake up automatically at midnight, run localized tests across my projects, scan for errors, and handle basic maintenance autonomously while I sleep.

Live voice transcription is another quality-of-life upgrade I have with Antigravity 2.0. Because it’s powered natively by the latest Gemini audio models, I don’t have to record a voice note, wait for an upload, and hope for a clean transcription.

Advanced slash commands

Quite helpful

Because Google optimized the application layer around Gemini 3.5, the execution speed is excellent. I ran some complex tasks on all three and Antigravity was faster than Codex and right up there with Claude Code in terms of generating code. The code quality was commendable too.

Google has also implemented advanced slash commands that let me direct the AI like an engineering manager.

Among them, I use the /goal command all the time. Here, I ask the agent to run without any hurdles, self-correct through errors, and bypass any need to approve actions from my side. It’s quite helpful for my small projects.

/grill-me is another command I use frequently. Instead of guessing my intent, the agent pauses and cross-examines me before touching a single file. /browser and /schedule are the other two commands you can use. I hope to see more such useful commands in future updates.

While Google is fully focused on Antigravity 2.0, the company hasn’t completely ditched the traditional IDE app. Of course, the company will eventually remove the Agent Manager from it and ask users to move to the 2.0 version, but if you are dealing with old projects and need a Gemini-powered text editor, you can always download the original app from the website.

Google's new AI coding beast

If Claude Code and Codex felt like the peak of the first generation of AI coding tools, Antigravity 2.0 feels like the arrival of the second. After all, developers don’t care about AI rivalries; we care about shipping clean code without the friction.

Whether you are trying to one-shot a complex microservice or seamlessly orchestrate a massive multi-file legacy codebase, one this is now undeniable: the bar for dev AI has just been raised, and the competition has some serious catching up to do.

Google Antigravity

Google Antigravity is an AI-powered IDE that rivals VS Code and Codex.