민프

[AI 뉴스] 구글 제미나이(Gemini) - 구글 Gemini, GPT-3.5 Turbo와의 비교 분석 본문

[정보]/[AI 뉴스]

[AI 뉴스] 구글 제미나이(Gemini) - 구글 Gemini, GPT-3.5 Turbo와의 비교 분석

민프야 2023. 12. 21. 12:00

최근 연구에 따르면, 구글의 AI 모델인 제미니(Gemini)가 OpenAI의 GPT-3.5 Turbo보다 성능이 떨어진다는 결과가 나왔습니다.

 


기사출처

https://venturebeat.com/ai/google-gemini-is-not-even-as-good-as-gpt-3-5-turbo-researchers-find/

 

Google Gemini is not even as good as GPT-3.5 Turbo, researchers find

The results are clearly a blow to Google's ambitions to go head-to-head with OpenAI in the generative AI race.

venturebeat.com

https://twitter.com/gneubig/status/1737108975039201533?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1737137149261025287%7Ctwgr%5E11cec1a90e8d294b551c816d702892e2bcbc9fe9%7Ctwcon%5Es3_&ref_url=https%3A%2F%2Fventurebeat.com%2Fai%2Fgoogle-gemini-is-not-even-as-good-as-gpt-3-5-turbo-researchers-find%2F

 

X의 Graham Neubig님(@gneubig)

TL;DR on the results? On all tasks, Gemini Pro achieved comparable but slightly lower accuracy than the current version of OpenAI's GPT 3.5 Turbo. Gemini and GPT were somewhat better than open-source contender Mixtral. But there’s quite a bit of nuance,

twitter.com

https://arxiv.org/pdf/2312.11444.pdf

 

Comments