Google has officially rolled out Gemini 2.5, which it claims is the most intelligent AI model its research arm, DeepMind, has developed for the Gemini AI chatbot. This latest iteration represents a significant leap in AI capabilities, enhancing reasoning, accuracy, and problem-solving.
Gemini was initially launched in December 2023 as Google’s response to OpenAI’s ChatGPT, which debuted in November 2022. The chatbot had been running on the Gemini 2.0 model, introduced in December 2024. Now, the introduction of Gemini 2.5 marks a new chapter in AI advancements.
Google describes Gemini 2.5 as a “thinking model,” designed to process complex reasoning before delivering responses. “Gemini 2.5 models are thinking models, capable of reasoning through their thoughts before responding, resulting in enhanced performance and improved accuracy,” said Google DeepMind CTO Koray Kavukcuoglu in a blog post.
Gemini 2.5 Pro, the first release in this series, is available through Google AI Studio and the Gemini app for Advanced users. Google also plans to integrate it into Vertex AI, its machine-learning platform, in the near future. Pricing details for scaled production use will be announced in the coming weeks.
The new model boasts significant improvements in reasoning, coding, math, and science benchmarks. Additionally, it can process vast datasets and interpret information from various sources, including text, audio, images, videos, and entire code repositories.
Currently, Gemini 2.5 Pro is exclusive to Gemini Advanced subscribers through the Ksh.3,700-a-month Google One AI Premium plan. This aligns with Google’s broader strategy of integrating AI advancements into its product ecosystem, including Search, Android, and Workspace.
The AI space is heating up, with Google facing competition from OpenAI, which offers a Ksh.26,000-a-month ChatGPT subscription for advanced research users. Other players include Elon Musk’s xAI, which integrates its chatbot Grok into the X social platform, and China’s DeepSeek, which has disrupted the industry by training its models with less advanced, lower-cost chips.
With the rollout of Gemini 2.5, Google is reinforcing its commitment to AI development, setting the stage for the next phase of intelligent computing.