Google announces Gemini 4 flagship AI model after months of delays

The introduction of Argon, described by a company spokesperson as larger and Google's "most performant model yet," suggests a push to advance its standing in the AI sector. Google claims Argon is comparable to competing frontier models on "key coding and cyber benchmarks," although a company press release noted it trailed on certain other metrics, including two of four coding benchmarks.

This release follows significant internal changes, including the cancellation of Gemini 3.5 Pro, which CEO Sundar Pichai had initially projected for June. The article notes that months of delays have caused Google to fall behind competitors. During this period, Google overhauled its DeepMind AI lab, with its founder and CEO stepping aside and several Gemini model leaders departing.

Google's messaging around its AI models appears to be evolving. While Pichai previously dismissed concerns about losing ground, the company has recently shifted focus to highlighting its cost advantage over competitor models. The current limited release of Argon to cybersecurity partners, without a public timeline, indicates a cautious approach to its broader deployment.