Google released Gemini 2.0 for the public on Wednesday. The general public can now access Gemini 2.0 Flash through the Gemini API in Google AI Studio and Vertex AI. The model, known as Google's most capable artificial intelligence (AI) model, was first released in December 2024. Initially, it was available only to developers and trusted testers.
Along with 2.0 Flash, Google has also released an experimental version of Gemini 2.0 Pro and a public preview of Gemini 2.0 Flash-Lite.
This follows the search engine's announcement of plans to invest $75 billion in expenses, including expanding its family of AI models this year. It is a significant increase from Google's $32.3 billion in capital expenditures in 2023, as the company races to keep up with AI competitors such as OpenAI, Microsoft, Meta, and Amazon-backed Anthropic.
Gemini 2.0 Flash
The Gemini 2.0 Flash model is designed for high-volume, high-frequency tasks at scale. It can perform multimodal reasoning across vast amounts of information with a context window of 1 million tokens.
This version has shown improved performance in key benchmarks such as MATH, SimpleQA, Bird SQL (Dev), and more. Google has also announced that an image generation and text-to-speech version will be released soon.
Gemini 2.0 Pro
Google claimed that Gemini 2.0 Pro has the best coding performance and the ability to handle complex prompts, along with enhanced understanding and reasoning of world information, making it the most advanced model they have produced so far.
The model features Google's largest context window of 2 million tokens, enabling it to fully analyse and comprehend massive volumes of data. It also has the capability to call tools such as Google Search and execute code.
Gemini 2.0 Pro is currently available as an experimental model for Google AI Studio and Vertex AI developers, as well as for Gemini Advanced users, via the model drop-down on both desktop and mobile.
Gemini 2.0 Flash Lite
Google announced that after receiving positive feedback on the price and performance of 1.5 Flash, it developed Gemini 2.0 Flash-Lite, a new model that offers higher quality than 1.5 Flash while maintaining the same speed and price. The Gemini 2.0 Flash-Lite outperforms 1.5 Flash in most benchmarks.
Like 2.0 Flash, it features a 1 million-token context window and supports multimodal input. For example, it can generate an appropriate one-line description for approximately 40,000 unique images for less than $1 in Google AI Studio's paid tier.
