October 6, 2026
Google launched its new model for artificial intelligence under the name “Gemini 4 Argon,” confirming that it is one of its most powerful models to date, with a special focus on completing long and complex tasks in the sectors of programming, cybersecurity, and specialized research.
The new model is distinguished by its superior ability to raise the output limit to one million tokens, compared to only 64 thousand tokens in the previous version, which allows it to carry out extended operations, and deal with multi-stage software projects and research within one continuous task. Google explained that it successfully employed it to correct errors, design algorithms, and transfer huge software databases between different languages.
The capabilities of “Gemini 4 Argon” are not limited to technical fields, but rather extend to specialized tasks in the financial, law, and tax sectors, in addition to its advanced ability to analyze charts and long video clips with high accuracy.
In terms of cybersecurity, Google paid special attention to the model’s ability to discover software vulnerabilities, verify them, and provide proposals to fix them. As a result of these sensitive capabilities, the company began making the model available in a limited and restricted manner to trusted testers and cybersecurity experts, in preparation for expanding its scope of use later.
Google has not yet revealed a specific date for the general and complete launch of the model, stressing that it will be made available to developers, companies, and users gradually in the coming period.
