Visual China Group | Getty Images
google As the search giant looks to align with ChatGPT creator OpenAI, the company is debuting its latest artificial intelligence model, Gemini 3.
New AI models will allow users to get better answers to more complex questions, “so you can get what you need with fewer prompts,” Alphabet CEO Sundar Pichai said in one of several blog posts published by Google on Tuesday.
Gemini 3 will be integrated into the Gemini app, Google’s AI search products AI Mode and AI Overovers, and enterprise products. The rollout began Tuesday for some subscribers and will be rolled out more broadly in the coming weeks.
This announcement comes about eight months after Google announced Gemini 2.5 and 11 months after Gemini 2.0. OpenAI, which kicked off the generative AI boom with the public release of ChatGPT in late 2022, introduced GPT-5 in August.
“It’s amazing to think that Al has evolved from simply reading text and images to reading a room in just two years,” Pichai wrote in one of his posts on Tuesday. “Starting today, we’re shipping Gemini at the same scale as Google.”
According to the company, the Gemini app currently has 650 million monthly active users and AI Overviews has 2 billion monthly active users. In August, OpenAI announced that ChatGPT had reached 700 million weekly users.
Pichai added that the latest model is “built to capture depth and nuance,” and said Gemini 3 is “much better at understanding the context and intent behind a request, so you can get what you need with fewer prompts.” Google’s other AI models may continue to be used for simpler tasks, the company said.
Alphabet and its giant rivals are spending millions to build infrastructure for AI development and quickly create more services for consumers and businesses. Alphabet, Meta, Microsoft and Amazon each raised their outlook for capital spending in their earnings reports last month, and together they expect that number to reach more than $380 billion this year.
In a statement from Demis Hassabis, CEO of Google’s AI arm DeepMind, Google said AI responses powered by Gemini 3 will “trade platitudes and platitudes for real insights, telling you what you need to hear, not what you want to hear.” Industry critics say today’s AI chatbots are too sycophantic.
Last week, OpenAI issued two updates to GPT-5. According to the company, one is “kinder-hearted, more intelligent, and better at following instructions,” and the other is “faster at simple tasks and more persistent at complex tasks.”
Google also announced a new agent platform called Google Antigravity that allows developers to code at a “higher task-oriented level.”
Josh Woodward, vice president of Google Labs and Gemini, told reporters at a briefing that Gemini 3 is the company’s “best-ever vibe coding model.” Vibe coding refers to the rapidly emerging market of tools that enable software developers to generate code with prompts.
Google says the new model will enable “generative interfaces” that provide specific answers in the form of a digital magazine. For example, the company asked Gemini to explain the Van Gogh Museum, including the life background of each work. The result was a colorful image-based description of each painting.
In AI mode, Gemini 3 will be offered to paying subscribers first. You’ll be able to analyze questions and create layouts using visual elements such as images, tables, and grids. Google said it can create custom-made interactive loan calculators and interactive simulations of complex physics problems.
Developers will have access to the Gemini API, and businesses will be able to integrate it through Vertex AI, a Google cloud service designed to build, deploy, and manage AI models.
For enterprise customers, the company says Gemini 3 can create employee onboarding and training, more accurately analyze videos and factory floor images, process procurement, and more.
Spotlight: Google adds Gemini to Chrome for everyone
