Google Opened Free Access To The Neural Network Gemini 1.5 Pro

Google has opened free access to its powerful neural network – Gemini 1.5 Pro – to everyone. One of the most outstanding features of this model is the context window, which allows you to process queries of up to 1 million characters. For example, Gemini can analyze several books and create a short translation from them.

Gemini 1.5 Pro works with a standard context window of 128,000 tokens, although the processing volume can reportedly be increased to 1 million tokens. At one time, Gemini 1.5 Pro is capable of processing up to an hour of video, 11 hours of audio, codebases with more than 30,000 lines of code or more than 700,000 words. During the study, Google also successfully tested processing up to 10 million tokens.

Using Transformer and MoE architecture, Gemini 1.5 combines the strengths of both models. Gemini 1.5 Pro excels at a variety of tasks, such as analyzing historical documents, such as the transcript of the Apollo 11 mission. A neural network is capable of not only analyzing large blocks of data, but also quickly finding a specific piece of text within them. Gemini 1.5 also handles large chunks of code efficiently. In the AI ​​Studio interface, the neural network is currently available with a limit of 20 requests per day.