Reasoning across a 402-page transcript | Gemini 1.5 Pro Demo

180 909
24.1
Google11.8 млн
Опубликовано 15 февраля 2024, 15:00
This is a demo of long context understanding, an experimental feature in our newest model, Gemini 1.5 Pro using a 402-page PDF transcript and a series of multimodal prompts.

This demo is a continuous recording of a live model interaction. Sequences have been shortened with response times shown.

Token count details: The input PDF file (326,658 tokens) and image (256 tokens) total 326,914 tokens. The text inputs add additional tokens into the prompt, yielding the 327,309 token total shown in the interface.

To learn more about Gemini 1.5, visit goo.gle/3weBZhn

Subscribe to our Channel: youtube.com/google
Tweet with us on X: twitter.com/google
Follow us on Instagram: instagram.com/google
Join us on Facebook: facebook.com/Google
автотехномузыкадетское