Google makes its Gemini chatbot faster and more widely available
Google has significantly upgraded its Gemini chatbot experience with the introduction of Gemini 1.5 Flash, a lightweight multimodal model that promises improved performance and wider availability. This update aims to keep pace with generative AI rivals like Anthropic and OpenAI. Gemini 1.5 Flash will be available on the web and mobile in 40 languages and around 230 countries, offering enhancements in quality and latency, particularly in reasoning and image understanding.
This new model is designed to be more efficient and cost-effective, which could benefit Google in terms of operational costs. Additionally, the context window of Gemini has been expanded to 32,000 tokens, equivalent to roughly 24,000 words or 48 pages of text. This allows the chatbot to summarize and reason over longer text snippets and files more effectively.
Another significant feature is the ability to upload files to Gemini for analysis, which was previously only available in the paid edition, Gemini Advanced. Users will soon be able to upload files from Google Drive and local devices, enabling tasks such as creating practice questions from study guides and analyzing data files to uncover insights and visualize them through charts and graphics.
To combat the issue of hallucinations, where the AI model generates false information, Google is introducing a feature that displays links to related web content beneath certain Gemini-generated answers. This aims to provide transparency on the sources of information used by the chatbot.
Gemini is also being rolled out to more users, including teenagers globally, and will be available in the European Economic Area, the U.K., and Switzerland, with support for additional languages such as French, Polish, and Spanish. The Gemini mobile app will also be launched in more countries.
