Google launches Gemini 4 Argon for solving complex real-world tasks
Google AI announced the global launch of its advanced model, Gemini 4 Argon. This model is designed to solve complex, multifaceted real-world problems, covering areas such as coding, finance, law, cybersecurity, and science.
Argon demonstrates impressive automation capabilities: for example, in the @antigravity tool, it can create custom hardware devices from a sketch, calculating mechanical compatibility and geometry in minutes, which reduces days of manual engineering work.
The model goes beyond simple text, acting as a multi-functional agent capable of automating business processes—from creating client presentations based on disparate data to turning ideas into full-fledged games.
For developers, Argon is available through the paid Gemini API tier, and for enterprise clients, it is available through Gemini for business and specialized tools.
Why it matters
- —This confirms the shift of AI from content generation to executing complex, multi-stage tasks (agency).
- —Demonstrating the model's application in highly specialized fields (CAD, jurisprudence) increases its value for the corporate sector.
- —Significantly reducing the time for engineering calculations (days to minutes) makes the model a breakthrough in industrial design.
Key facts
- Gemini 4 Argon is an advanced model for complex tasks in coding, finance, and jurisprudence.
- The model can generate custom hardware devices from a 2D sketch, calculating mechanical parameters.
- Argon is capable of automating business processes: from data collection to creating presentations and planning subsequent actions.
- Available for developers via the paid Gemini API and for corporations via Gemini for business.
The full text is in the original source. Here we provide a brief summary and key facts.