About the news
According to Google’s official announcements, the model is now accessible in these places:
- Developers and enterprises can use it through the Gemini API and Google AI Studio
- Google AI Ultra subscribers can access it in Google Antigravity (the agent-first development platform)
- Google AI Pro and Ultra subscribers can use it in the Gemini app and in AI Mode in Search
Google positions Gemini 4 Argon as a frontier model built for complex, long-horizon workflows. The company highlights its strengths in real-world software engineering, enterprise knowledge work (including legal and finance), and cybersecurity defense.
Key points from the official materials:
- The model is designed for multi-step problems that require sustained reasoning and structural precision, such as large codebase migrations and frontier research
- Google reports that Argon agents have already been used internally on demanding engineering tasks, including performance optimizations and large-scale code migrations
- Introductory API pricing was listed at $2 per million input tokens and $10 per million output tokens, with cached input tokens offered at a significant discount
- Company-reported benchmark results place the model competitively on several knowledge-work and agentic coding evaluations
Why it matters
Google has also published supporting materials on safety and security for the model, describing the evaluation approach and intended use cases. The public availability expands access beyond the earlier trusted-partner and limited programs.
In short, Gemini 4 Argon moved from restricted early access into wider developer and subscriber availability, with a clear focus on long, complex professional workflows rather than short single-turn queries.
