中国科技公司元象近日宣布推出一款名为XVERSE-Long-256K的开源大模型，该模型支持上下文窗口长度达到256K，能够处理高达25万汉字的输入，标志着大模型应用正式迈入“长文本时代”。值得一提的是，这款模型不仅全开源，而且无条件免费商用，元象还贴心地提供了手把手训练教程，使得广大中小企业、研究者和开发者得以尽早实现“大模型自由”。
In a landmark move, Chinese tech firm Yuanxi has launched a groundbreaking open-source large model, XVERSE-Long-256K. This model supports a context window length of up to 256K, enabling it to process inputs of up to 250,000 Chinese characters, heralding the arrival of the “long-text era” for large model applications. Notably, this model is not only open-source but also available for commercial use at no cost, and Yuanxi has generously provided step-by-step training tutorials, allowing a vast number of SMEs, researchers, and developers to achieve “large model freedom” sooner.
【来源】https://mp.weixin.qq.com/s/R8ewi1NsAK9Qwh0e7UyyBw
Views: 1