Google has announced the launch of Gemini 2.5, its latest and most advanced artificial intelligence model, promising a significant leap in AI reasoning and performance.
This release comes amidst an increasingly intense competition within the AI sector, with notable advancements from players like Chinese AI firm Deepseek and industry heavyweight OpenAI.
The company claims the 2.5 is its most intelligent AI model, adding that the experimental version of Gemini 2.5 Pro has achieved state-of-the-art results across a range of benchmarks, notably topping the LMArena leaderboard, a measure of human preferences, by a “significant margin.”
Smarter AI
According to Google’s official announcement, Gemini 2.5 models are “thinking models,” designed to reason through their thoughts before generating responses.
- This approach, Google asserts, leads to enhanced performance and improved accuracy, moving beyond simple classification and prediction to incorporate complex analysis, logical conclusions, contextual understanding, and informed decision-making.
- Google noted that for a long time, it had explored ways of making AI smarter and more capable of reasoning through techniques like reinforcement learning and chain-of-thought prompting.
“Building on this, we recently introduced our first thinking model, Gemini 2.0 Flash Thinking.
“Now, with Gemini 2.5, we’ve achieved a new level of performance by combining a significantly enhanced base model with improved post-training. Going forward, we’re building these thinking capabilities directly into all of our models,” the company stated.
Key highlights of Gemini 2.5 Pro
According to Google, the Gemini 2.5 Pro Gemini 2.5 Pro is state-of-the-art across a range of benchmarks requiring advanced reasoning.
- Without test-time techniques that increase cost, like majority voting, 2.5 Pro leads in math and science benchmarks like GPQA and AIME 2025.
- It also scores a state-of-the-art 18.8% across models without tool use on Humanity’s Last Exam, a dataset designed by hundreds of subject matter experts to capture the human frontier of knowledge and reasoning.
- Google added that the 2.5 Pro excels at creating visually compelling web apps and agentic code applications, along with code transformation and editing. On SWE-Bench Verified, the industry standard for agentic code evals, Gemini 2.5 Pro scores 63.8% with a custom agent setup.
Gemini 2.5 Pro is currently available in Google AI Studio and the Gemini app for Gemini Advanced users. Google has also announced that it will be available on Vertex AI in the coming weeks. Pricing for scaled production use with higher rate limits will be introduced shortly.
What you should know
Recently, Chinese AI firm Deepseek has made notable strides in developing efficient and powerful reasoning models, putting pressure on established Western AI companies. Their focus on open-source models has also created a competitive environment.
- OpenAI continues to push the boundaries of AI capabilities, with ongoing updates to its GPT models.
- The competition between Google and OpenAI, in particular, remains a central narrative in the AI industry.
