
The goal of Google’s most recent artificial intelligence model, Gemini 2.5, is to improve cognitive abilities in mathematical problem-solving, scientific thinking, and software creation. The model, which was released on March 25, is now accessible to developers in its Pro edition, establishing itself as Google’s most sophisticated AI to date.
Known as a “thinking model,” Gemini 2.5 is intended to analyze issues before coming up with a solution, resulting in more precise and contextually aware outputs, especially in high-precision fields like code creation and applications connected to STEM. Citing top-tier ranks in the technical, knowledge, and reasoning domains, the tech giant highlighted the model’s exceptional performance in internal benchmarks.
Gemini 2.5 placed second in the real-time WebDev leaderboard, a performance metric that assesses AI models in web development jobs, according to LMArena, an AI benchmarking and competition platform. Gemini 2.5 beat competitors including DeepSeek, Grok, and OpenAI’s ChatGPT with an arena score of 1267.70. Claude 3.7 Sonnet, created by Anthropic, took first place with a score of 1354.01, but it was behind it.
In the larger market for AI products, Google still lags behind OpenAI despite this technological advancement. With over 5.2 billion monthly visitors, OpenAI’s flagship model, ChatGPT, maintained a dominant 43% market share in February 2025. In 2024, it received over 40 billion visits annually, confirming its dominance in the market.
According to data from the AI analytics company aitools.xyz, 101 billion visits were made to the AI tools industry overall in 2024. 10.4 billion visits, or 10.25% of the market, gave Canva’s AI generator the second-highest share. Thanks to 792 million monthly visits, upstart competitor DeepSeek has risen quickly, taking third place with a 6.58% share and growing at a rate of 195% in recent months.
The AI industry is about to enter a phase of rapid innovation and strategic positioning as competition heats up. Although Google’s Gemini 2.5 represents a major advancement in model sophistication, the disparity in market uptake highlights OpenAI’s ongoing supremacy in applications aimed at consumers.
In the upcoming months, industry observers will be keenly monitoring if Google’s improved performance benchmarks and developer accessibility can result in wider acceptance.