Google Introducing GEMINI

on

Google just dropped it's ChatGPT killeršŸ”„


Google says Gemini beats GPT4 in 30 out of 32 benchmarks.


Here are 7 things you should know about:


1. Multimodality

Google's new AI model, Gemini can understand and combine:

• Text

• Code

• Audio

• Images

• And video.


2. Advanced Reasoning:  

ChatGPT can’t solve complex maths but Gemini can. 

Process complex written and visual information -Works in multiple fields including Science, Maths and Finance


3. Coding Capabilities:  

A specialized version of Gemini, Alphacode2 - Performs better than 85% of competition participants - up from nearly 50% for AlphaCode. 

Understand, Explain and Generate high-quality code.


4. Sizes and Availability:  

3 sizes of Gemini AI model Ultra for complex tasks, rolling out early next year. 

Pro for scaling, available today in Google Bard. 

Nano for on-device tasks, available on Pixel 8 Pro.


5. Development

Starting on December 13, developers and enterprise customers can access Gemini pro via the Gemini API in Google AI studio or Google cloud Vartex AI.

Google further refining the gemini Ultra moder befor emaking it broadly available early next year.


6. Gemini benchmarks  

In terms of performance, Google showed Gemini Ultra surpassing GPT-4 in text-based benchmarks that measure reasoning, math, and code.



7. Bard with Gemini Pro  

Bard with Gemini Pro, a new version of Google's AI model,offers more advanced reasoning, planning, writing, and content understanding. 



It's available now.

bard.google.com


Comments

Post a Comment