Gemini 4 Argon: Who Can Access It and What It Costs
Gemini 4 Argon is announced but not broadly available. Current access is limited to trusted cyber defenders, with paid API customers and Google AI Ultra subscribers named as the first future groups.
Direct answer: Gemini 4 Argon has been announced, but it is not broadly available in the consumer Gemini app or to ordinary developers yet. Google says initial external access is limited to trusted cyber defenders through its Fairwind Program. Broader access will come later, starting with paid API customers and Google AI Ultra subscribers, but Google has not published an exact release date.
What did Google announce?
Google introduced Gemini 4 Argon on September 30, 2026 as a frontier model built for long, complex work across software engineering, enterprise knowledge tasks, multimodal analysis and cybersecurity defense. The announcement arrived at 20:00 UTC, which was October 1 in Asia/Shanghai.
Google says Argon is already used in internal workflows and is being released first to a limited group of trusted cyber defenders. This is a staged launch, not a general Gemini app or developer API release.
Who can use Gemini 4 Argon now?
The current external group is a set of trusted cyber defenders in Google's Fairwind Program. Google also describes internal use by its own teams. The company does not provide a public signup path, a public API model ID or a complete eligibility list in the launch post.
Google says broader availability will begin with paid API customers and Google AI Ultra subscribers before expanding to developers, enterprises and consumers. “Rolling out soon” is not a calendar date, so readers should not assume that an existing Gemini subscription or API account already includes Argon.
How much will Gemini 4 Argon cost?
Google announces introductory pricing of $2 per million input tokens and $10 per million output tokens. Cached input tokens are stated as 95% below the input-token price. After the introductory period, Google says pricing will rise to $4 per million input tokens and $20 per million output tokens.
The launch post does not state how long the introductory period lasts. Pricing also does not prove that an account can access the model today; availability, model identifiers, regions, quotas and product packaging still need to be confirmed when Google opens each release stage.
What is different about the model?
Google highlights a 1-million-token output limit, up from the 64,000-token output limit it cites for the prior generation. The company positions that capacity for long-running reasoning and work that may require hundreds of thousands of generated tokens in one trajectory.
Google also publishes results for coding, automation, finance, legal and long-video evaluations, plus examples from internal engineering work. These are useful details about Google's launch claim, but they are company-reported results. They should not be treated as independently reproduced proof that Argon is best for every workload.
Why is access limited?
Google says the phased release gives it time to strengthen safeguards before broader availability. The launch specifically discusses preventing misuse, resisting indirect prompt-injection attacks, monitoring for misaligned behavior and hardening secure evaluation environments.
The company also says it is participating in the United States government's voluntary process for pre-release model access. That explains the initial trusted-tester boundary; it does not supply a guaranteed public launch date.
What can ordinary Gemini users use today?
Gemini 4 Argon is separate from the current Gemini app change that replaces Gems with reusable Skills. Skills are already rolling out in Gemini chat, while Argon remains a staged frontier-model release. See Erawish's Gemini Gems-to-Skills migration guide for the consumer feature, account and migration timeline.
What should you check before planning around Argon?
- Access: Wait for an official account-level availability notice rather than assuming a paid plan is already eligible.
- Price: Confirm whether introductory or later pricing applies when the model actually appears in your product or API account.
- Limits: Check the published model ID, regions, quotas, context limits and output limits at launch; only the 1-million-token output claim is stated in this announcement.
- Evaluation: Test the model on your own representative tasks and safety requirements instead of treating launch benchmarks as a universal production result.
Frequently asked questions
Is Gemini 4 Argon in the Gemini app now?
Google's launch post does not say that ordinary Gemini app users have access now. It says trusted cyber defenders receive initial access and names Google AI Ultra subscribers as one of the first future groups.
Can developers call Gemini 4 Argon through the API now?
The announcement names paid API customers as an early future group, but it does not publish a general API availability date or public model ID. Developers should wait for current Google API documentation and account-level access.
Does the announced price mean the model is publicly available?
No. Google can announce future pricing before broad access opens. The launch terms explain expected token costs, while the release remains staged.