Google CEO Sundar Pichai announced that Gemini 4 Argon, the company’s flagship model unveiled on September 30, is now opening up to a much wider set of users.
“It’s available globally today for developers and enterprises in the paid tier of the Gemini API and the Gemini platform, for Ultra subscribers in Google Antigravity, and Ultra and Pro subscribers in the Gemini app and Search,” Pichai wrote in a post announcing the rollout.
Google’s own release notes flesh out the access tiers further. Developers and enterprises get the model globally through the Gemini API’s paid tier and AI Studio, with enterprises also able to reach it through Gemini for business. Google AI Ultra subscribers get it in Antigravity, AI Studio and Android Studio, while both Ultra and Pro subscribers can use it in the Gemini app and AI Mode in Search. Separately, cybersecurity defenders can still apply to the Fairwind Program for a version of the model without cyber guardrails, and Google says it’s opening applications for 10,000 scientists and researchers to get six months of free access to Antigravity’s highest usage tier, subject to eligibility and capacity.
The expansion marks a significant shift from how Argon was initially released. When Google first introduced the model, access was restricted to a small group of trusted cyber-defense partners through the Fairwind Program, alongside the US government, with no public timeline for a broader launch. Developers, enterprises and everyday Gemini users were left waiting while the model was evaluated internally. With this update, that waiting period is over for a large chunk of Google’s user base — paying subscribers can now tap into Argon directly from the surfaces they already use daily, whether that’s the Gemini app, Search, or Google’s agentic coding environment, Antigravity.
Google has positioned Argon as a model built for complex, multi-step work across coding, finance, legal, cybersecurity and science, rather than one narrowly focused on chatbot-style conversations. The company says it ranks first on the Vals Index, which measures performance across economic sectors like finance, coding, legal and tax, and scores 87% on the Embodied Reasoning (ERQA) benchmark that tests physical and spatial reasoning. It also ships with a 1 million-token unified input-output limit through the Gemini API, backed by what Google calls a new continuation protocol for long outputs — a sharp jump from the 64,000-token ceiling on its predecessor, which the company says helps it handle large code migrations and long-running tasks in a single run. Google also claims the model writes with less hedging and sticks to style instructions better than earlier Gemini releases, and that it’s more resistant to prompt injection and less likely to fulfil harmful requests than previous models.
On pricing, developers get Argon at an introductory rate of $2 per million input tokens and $10 per million output tokens, with cached input tokens discounted 95%. Those rates hold until November 30, 2026, after which the public-access model moves to $4 per million input tokens and $20 per million output tokens.
The rollout comes at a tense moment for Google’s AI division, after an executive reshuffle and a months-long delay that saw the company quietly scrap plans for a previously promised Gemini 3.5 Pro release. By tying availability to its Pro and Ultra subscription tiers, Google is giving paying users a clear incentive to upgrade, while developers and enterprises get access through the paid API tier rather than the free playground. The gating to Ultra subscribers specifically for Antigravity suggests Google sees the combination of Argon’s agentic strengths and its coding IDE as a premium offering aimed at power users and professional developers rather than casual coders.
Gemini’s broader position in the market has been under pressure in recent months, with the product recently losing website traffic share for a fourth straight month as ChatGPT clawed back ground. Argon’s benchmark standing has also been a mixed bag — it trades wins and losses against rivals depending on the test, and newer entrants have continued to crowd the leaderboard, with Meta’s Muse Wonder recently landing third on the Artificial Analysis Intelligence Index, ahead of both GPT-6 Astra and Gemini 4 Argon. Whether wider access to Argon through the Gemini app, Search and Antigravity can help Google claw back some of that ground will likely depend on how the model performs for subscribers in everyday use, rather than on benchmark charts alone.