StepFun releases "trillion" and "multi-mode" large models

StepFun large models

DIYuan | 2024-07-04 17:20

【數(shù)據(jù)猿導讀】 StepFun releases "trillion" and "multi-mode" large models

On July 4, at the 2024 World Artificial Intelligence Conference, StepFun released three new Step series general large models: STEP-2 trillion parameters language large model official version, STEP-1.5V multimodal large model, STEP-1X image generation large model. Among them, Step-2 greatly improves the training efficiency and approximates GPT-4 in the aspects of mathematics, logic, programming, knowledge, creation, multi-round dialogue, etc. Step-1.5V comprehensively improves the image perception and understanding ability, and has a strong video understanding ability. At the same time, StepFun also focused on showing the self-developed large-scale model application product intelligent assistant "Leap Ask" and the AI open world platform "Bubble Duck" for C-end users, and disclosed the latest progress and plans in the field of large-scale model ecological cooperation.

來源：DIYuan

收藏分享

聲明：數(shù)據(jù)猿尊重媒體行業(yè)規(guī)范，相關(guān)內(nèi)容都會注明來源與作者；轉(zhuǎn)載我們原創(chuàng)內(nèi)容時，也請務必注明“來源：數(shù)據(jù)猿”與作者名稱，否則將會受到數(shù)據(jù)猿追責。