Google Unveils Gemini 4 Argon AI Model, Shares Rise in After-Hours Trading

Google on Wednesday unveiled Gemini 4 Argon, its next-generation flagship artificial intelligence model, in a move that boosted parent company Alphabet's (GOOGL) shares by approximately 2% in after-hours trading. The announcement, which follows weeks of anticipation, positions the new model as a direct competitor to leading systems from OpenAI and Anthropic.

A Preview of New Capabilities

In a post on the social media platform X, Google CEO Sundar Pichai introduced Argon as a model demonstrating "frontier performance" in complex tasks related to software engineering, cyber defense, and other advanced workflows. The release is currently limited, with Reuters reporting that Argon is initially being provided to a select group of cybersecurity partners.

Google has not yet provided a timetable for the model's general public availability. The announcement serves as an early look at the system's capabilities, following comments last week from a company executive who indicated a version would be released "as soon as possible."

Competitive Benchmark Results

Google released a suite of benchmarks positioning Argon ahead of its primary competitors, OpenAI's GPT-6 Astra and Anthropic's Claude Opus 5.5, on most of the tests shown. While the results did not represent a complete sweep, Argon demonstrated notable strengths.

Key performance metrics published by Google include:

Ad

• Vals Index: Argon scored 68.9%, compared to 67.0% for Claude Opus 5.5 and 63.1% for GPT-6 Astra.
• Coding: The model led on the DeepSWE v1.1 and Vibe Code Bench tests, scoring 77.9% and 91.9% respectively.
• Long-Context: Argon showed a significant advantage, achieving 99.7% accuracy on the GraphWalks test with up to 128,000 tokens and 84.2% on tests with up to 1 million tokens.
• Complex Reasoning: It scored 39.5% on the Agent’s Last Exam benchmark, ahead of GPT-6 Astra’s 34.2%.

Competitors did outperform Argon on several individual tests, with GPT-6 Astra leading on FrontierSWE v2 and Claude Opus 5.5 posting the top score on Terminal-bench 4.0.

Market Context and Outlook

The unveiling of Argon is a critical step for Google as it seeks to maintain its competitive edge in the rapidly advancing AI sector after experiencing product delays. The strong benchmark performance is intended to signal Google's ability to compete at the highest level.

However, the provided benchmarks come amid questions about the model's real-world performance. A recent Bloomberg report, citing people with direct access, suggested that while Gemini 4 performed well on industry benchmarks, it has been less impressive in internal, real-world applications. Google has pushed back on this characterization, highlighting its internal use of Argon for complex tasks such as coding and data-center optimization.