Google has introduced Gemini 4 Argon, its latest frontier AI model designed for complex, long-horizon workflows across software engineering, enterprise knowledge work, finance, legal applications, and cybersecurity defense.
Google is initially rolling out Argon to a group of trusted cyber defenders through its Fairwind Program while taking a phased approach to broader access. The company is also participating in the U.S. government’s voluntary process for pre-release model access and plans to expand access to developers, enterprises, and consumers after further testing of its safeguards.
Argon will launch with introductory pricing of $2 per million input tokens and $10 per million output tokens, while cached input tokens will receive a 95% discount from the standard input-token price.
Google said thousands of employees are already using Argon internally for specialized coding, research, and writing tasks. In one quantum computing application, the model improved on a published baseline by 40% within minutes.
A team of Argon agents also analyzed Google data center telemetry to identify memory optimizations expected to free more than 300 TiB initially, with estimated total savings of 500 TiB to 1 PiB.
Argon agents are additionally being used to help migrate large C and C++ codebases to Rust, including projects ranging from tens of thousands of lines to more than 800,000 lines for the Fuchsia Zircon kernel. In another project involving Google’s libgav1 video decoder, Argon replaced 32,000 lines of SIMD code and produced a Rust implementation that Google said ran 2.7 times faster than the previous Rust port.
The model’s output limit has increased from 64,000 to 1 million tokens, giving Argon substantially more capacity for long-running reasoning and generation tasks.
On Google’s reported benchmarks, Argon scored 77.9% on DeepSWE v1.1 for long-horizon software engineering tasks and 51.3% on AutomationBench, where it ranked first. Google also reported leading performance across financial, legal and other enterprise evaluations.
