Google Opens Gemini 4 Argon to Developers and Businesses Worldwide After Restricted Debut

Google expanded access to Gemini 4 Argon, its most powerful AI frontier model to date, making it available to developers and businesses globally after an initial restricted rollout. The broader release follows an earlier debut that had limited Argon to trusted cybersecurity specialists through Google's Fairwind Program. Alphabet shares fell roughly 2% on the initial launch, with investors concerned about the narrow early rollout and what it meant for near-term enterprise revenue.

Gemini 4 Argon is designed for demanding professional tasks including software engineering, cybersecurity, and complex scientific and financial work, with Google citing major improvements in coding and deep reasoning. The model supports a context window of up to 1 million tokens per run, a significant expansion from the previous 64,000-token limit, allowing it to sustain far longer reasoning chains without interruption. Google set introductory pricing at $2 per million input tokens and $10 per million output tokens, with standard pricing expected to roughly double after the introductory period, putting additional competitive pressure on OpenAI and Anthropic.