Alphabet-owned Google has announced a new flagship artificial intelligence model to lead its Gemini 4 generation, ending months of delays as the company seeks to strengthen its position against rivals OpenAI and Anthropic.
The new model, called “Argon”, is larger than Google’s previous advanced Pro models, according to a company spokesperson.
“It’s our most performant model yet built for complex workloads,” the spokesperson said, adding that Google considers Argon comparable to frontier models such as OpenAI’s Astra and Anthropic’s Opus on selected coding and cybersecurity benchmarks.
Argon focused on advanced workloads
Google said Argon is being made available to selected cybersecurity partners as part of its efforts to assess the model’s capabilities and risks.
The company is also participating in the Trump administration’s voluntary process for providing access to AI models before their public release.
Google did not provide a timeline for when Argon would become available to the general public.
The company said the model has been designed to refuse requests that could facilitate cyberattacks or assist in the development of chemical, biological or nuclear weapons.
Mixed benchmark results
According to a Google press release, Argon recorded stronger self-reported results than Astra and Opus on several industry benchmarks.
However, the company acknowledged that the model remained behind its rivals on some other measures, including two of the four coding-related benchmarks included in Google’s release.
The mixed results highlight the competitive nature of the AI model race, where companies increasingly compete on coding, cybersecurity, reasoning and other complex tasks.
Google Tests New Feature to Move ChatGPT Users to Gemini AI
Google abandons Gemini 3.5 Pro
Google also confirmed that it no longer plans to release Gemini 3.5 Pro.
CEO Sundar Pichai had originally indicated that the model would be released in June, but the company has since changed its plans.
The delays have coincided with significant changes within Google’s AI organisation.
Google overhauled its DeepMind AI lab during the intervening period, while DeepMind founder and CEO Demis Hassabis stepped aside from his leadership role and several senior figures involved with Gemini left the company.
Google seeks to regain momentum
The prolonged delays have come as Anthropic and OpenAI continued releasing new versions of their leading AI models, intensifying competition at the top end of the industry.
Pichai pushed back in July against suggestions that Google was losing ground in artificial intelligence.
However, Google has increasingly shifted its messaging in recent months, placing greater emphasis on Gemini’s cost advantages compared with competing models rather than focusing exclusively on claims of cutting-edge performance.
With Argon now positioned as the flagship of the Gemini 4 generation, Google is seeking to regain momentum in a rapidly evolving AI market dominated by increasingly capable frontier models.
