AI model: Google Gemini 4 Argon

Gemini 4 Argon is now available for everyone

Overview: Google launched Gemini 4 Argon, its latest frontier AI model built for deep work and complex real-world tasks across fields like coding, finance, legal, cybersecurity, and science.

Performance & Pricing: It ranks #1 on the Vals Index and launches at an introductory price of $2 per million input tokens and $10 per million output tokens (with cached input tokens at a 95% discount) alongside a 1 million unified input-output token limit.

Key Capabilities:

Engineering & Code: Helped Google engineers increase agentic code submissions by over 35% while reducing rollbacks, tackling massive tasks like refactoring Chrome's legacy codebase.

Multimodal & Spatial: Demonstrates advanced spatial reasoning, physics-ready 3D asset creation, and video-to-simulation-to-real workflows for robotics.

Science & Research: Accelerates breakthroughs in areas like quantum computing error correction, nuclear fusion, rare disease diagnosis, and enzyme design.

Availability: Now rolling out globally via AI Studio, the Gemini platform, Google Antigravity, and the Gemini app for eligible subscribers and developers.