Google Launches Gemini 4, Mixed Internal Reviews on Programming Performance

Google has begun rolling out its flagship AI model, Gemini 4 Argon, but there are internal doubts about its performance in key areas such as programming. On Wednesday, Google opened the model to a small group of trusted cybersecurity partners and plans to expand its use after further testing, initially to paid users. Gemini 4 has excelled in several benchmark tests, including outperforming OpenAI's Astra model in a security capability test. However, some insiders believe these metrics do not fully reflect real-world performance. Sources indicate that while Gemini 4 performs well in benchmark tests, its effectiveness is not ideal during actual employee use, particularly in certain programming tasks.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.