Google is making waves in the world of artificial intelligence with its flagship suite of generative AI models, apps, and services known as Gemini. Developed by Google’s AI research labs DeepMind and Google Research, Gemini comes in three flavors: Ultra, Pro, and Nano. These models are trained to be “natively multimodal,” meaning they can work with more than just words, including audio, images, videos, and codebases in different languages.
Gemini is distinct from the Gemini apps, which serve as an interface for accessing the Gemini models. The models can perform a range of tasks, from transcribing speech to generating artwork. Gemini Ultra, the most performant model, can help with tasks like physics homework and identifying scientific papers relevant to a problem. Gemini Pro, an improvement over Google’s LaMDA, excels in reasoning and understanding capabilities. Gemini Nano, a smaller version, powers features on mobile devices like the Pixel 8 Pro and Samsung Galaxy S24.
Google claims that Gemini exceeds current state-of-the-art results on academic benchmarks and is more capable at tasks like summarizing content and brainstorming than its competitors. However, early impressions of Gemini Pro have been mixed, with users pointing out errors and struggles with translations.
While Gemini 1.5 Pro is currently free to use in the Gemini apps, AI Studio, and Vertex AI, it will cost $0.0025 per character once it exits preview in Vertex AI. Pricing for Gemini Ultra has yet to be announced.
Developers and users can try Gemini Pro and Ultra in the Gemini apps, Vertex AI, and AI Studio. Google has integrated Gemini models into various dev tools and security products, showcasing the versatility and potential of its AI technology.