Google releases Gemini 4 Argon, its new flagship model for developers and businesses

Google has begun rolling out Gemini 4 Argon, which it pitches at long, complex work such as coding, finance, legal and security tasks. Reports differ on the exact launch day.

Google releases Gemini 4 Argon, its new flagship model for developers and businesses

Google has released Gemini 4 Argon, which it calls its latest flagship AI model, to developers and businesses worldwide. According to Reuters, the company is trying to regain ground against rivals OpenAI and Anthropic.

The timing is described differently in different reports. A Reuters dispatch dated 13 October says Google released the model on Wednesday. A newsletter summary says it became globally available on Tuesday. The dates have not been reconciled here.

Advertisement

What Google says it does

In its announcement, Google says Argon is built for “deep work” on complex, real-world tasks. It names coding, finance, legal, cybersecurity and science as areas where the model is strongest. Google also says the model’s depth and precision on multi-step tasks set it apart from other models. These are the company’s own claims, and this article has not seen independent test results.

A newsletter summary describes the pitch as aimed at long, complex projects. That fits Google’s emphasis on multi-step tasks rather than quick chat replies.

In the Gemini app

Consumers are getting it too, but only some of them. Android Police reports that Gemini 4 Argon is rolling out in the Gemini app for paid subscribers, on the Pro and Ultra plans.

Advertisement

The same outlet says the model is meant to cut down on padding in answers. It describes response options such as “show me the facts,” a simplified explanation and a lighthearted recap. Details on how these options work, and on pricing for developers, were not available in the material reviewed for this article.

The competitive backdrop

Google has been under pressure to show that its newest model can match or beat the latest releases from OpenAI and Anthropic. Reuters frames the launch in those terms. Recent weeks have brought a run of new models from rivals, and each release has raised the bar that Google is measured against.

Whether Argon closes that gap will depend on independent benchmarks and on how developers and businesses use it in practice. Google’s announcement lists strengths but, in the portions reviewed here, does not settle how it compares directly with competing models.

Advertisement

Argon is the first model in the Gemini 4 line to be described this way in the reports reviewed. Further details, such as context length, pricing tiers and safety documentation, should appear in Google’s developer materials.

Sources