Gemini 4 Argon is now rolling out to Google’s Antigravity for AI Pro subscribers

Google is now presenting assistance for Gemini 4 Argon to Antigravity, its AI coding tool.

Gemini 4 Argon is Google’s most current flagship AI design, revealed at the end of September and having actually extensively presented to users previously today. Now, it’s likewise offered within Antigravity.

Antigravity is an agentic advancement platform, as Google explains it, which leverages AI and representatives to help in coding and developing apps. Google revealed that Argon would be readily available in Antigravity previously today, and it’s now actively presenting to AI Pro and AI Ultra customers– we’re seeing it reside on our end, too.

In a post on Twitter/X, Google includes that there’s “a different weekly-only Argon quota” on top of the existing quota for Gemini 3.8 Flash, suggesting you’ll have the ability to utilize the design that finest fits your requirements. Google states that Argon is “developed to tackle your hardest issues.”

Google likewise supplied a host of examples of how Argon in Antigravity has actually been utilized internally, consisting of:

• YouTube remark optimization: Argon in Antigravity assisted the group cut the CPU footprint of YouTube’s remarks frontend in half. By tracing production profile information, Argon determined that redundant emoji making alone represented almost 45% of server overhead, causing 12 targeted optimizations. The updates lowered server comment-load latency by 30 ms and dropped customer end-to-end latency by 20 ms.
• Debugging Wing drone physics simulator: Argon in Antigravity evaluated Wing’s exclusive internal drone physics simulator, thinking through the underlying physics and mathematics instead of simply the structure of the code. It highlighted locations for enhanced physics simulations and proposed 18 targeted modelling enhancements throughout aerodynamics, propulsion, and structural characteristics to deal with enduring tradition concerns.
• Burning down the stockpile: Argon in Antigravity took an aging line of 134 open bug reports covering numerous services, and ran representatives to exercise which were still genuine, cross-referencing dedicate histories, inspecting whether prospect repairs really reached production, and checking live telemetry for whether the mistake still takes place. In under 3 hours it closed 97 stagnant or replicate bugs with recorded devote evidence (a 72.4% line decrease) and arranged the staying 37 by owner without any unassigned tickets left.
• Moving tradition code: For years, crucial clinical research study has actually been locked inside delicate tradition codebases like Fortran. Utilizing a custom-made ability within Antigravity, a group released Argon to map and perform a total reword into modern-day Rust. Argon autonomously provided the up-to-date codebase, accomplishing a 1.8 x efficiency increase and memory security while offering researchers a robust structure for future work.
• Carrying out security audits: Argon in Antigravity was utilized to investigate a batch of access-control modifications. Argon traced how authorizations in fact streamed at runtime instead of simply checking out the code diff, and discovered 4 important security bugs. 2 earlier frontier designs had actually investigated the similar code and both concluded there were absolutely no security holes.
• Triaging production assistance tickets: Argon in Antigravity batch-processes lots of tickets from an on-call rotation. It sorts them by problem type, tracing source through the code and logs, proposing repairs and preparing the replies for human approval. This has actually maximized 2-3 hours per shift, allowing the group to concentrate on dealing with concerns quicker.
• Marketing site localization: The Antigravity DevRel group utilized Argon in Antigravity to localize numerous pages of designer paperwork into 17 languages in a matter of hours. This enhanced the ease of access and UX of the site by diminishing a multi-week procedure to hours and moving human effort solely to premium evaluation.
• Speeding up information center pipeline modernization: Showing how AI-augmented execution assists groups do more with less, a Data Science Technical Program Manager leveraged Argon in Antigravity to fast-track advancement of a 3,200-line Google Data Center fleet analytics pipeline (Python, SQL, and control panels) into production in days. Coupling with Argon to quickly deal with complicated multi-system edge cases, remove hardcoded tradition possessions, and enhance timeseries questions versus timeouts drove end-to-end performance– providing a tidy, production-grade, and extensible architecture for fleet-scale analytics.
• Vetting sales potential customers: Argon in Antigravity developed an automated certification workflow for an account executive in Ads sales that checks potential business site traffic, validates the website has clear conversion funnels, and cross-references the sales database to see whether the account is a competent possibility. This lowered research study time by 90%, conserving the account executive upwards of 10 hours of manual labor weekly so they can hang around on direct customer outreach rather.
• Financial preparation and analysis: With custom-made financing plugins, a financing lead utilized Argon in Antigravity for end-to-end monetary reconciliations, budget plan tracking, agreement extraction, and day-to-day executive instructions throughout 80 different workflows. Over 7 organization days, 15.5 hours of human instructions produced about 335 human-equivalent hours of confirmed monetary output, conserving more than 8 weeks of manual labor.

More on Gemini:

• Google Antigravity Googlebook app strikes the Play Store
• Google presenting Universal Cart that lets you go shopping the web and discover handle Gemini, Search
• Gemini AI Plus customers will lose access to Pro designs next month

Follow Ben: Twitter/XThreadsBlueskyand Instagram

Kyle Bradshaw added to this post.