OpenBMB

The large-scale pre-trained language model library and related tools supported by the Tsinghua team.
The full name of OpenBMB is Open Lab for Big Model Base, which aims to build a large-scale pre-trained language model library and related tools, accelerate the training, fine-tuning and reasoning of large models of more than 10 billion levels, reduce the threshold for using large models, and cooperate with developers at home and abroad Strive to form an open source community of large models, promote the ecological development of large models, realize the standardization, popularization and practicality of large models, and let large models fly into thousands of households.
The OpenBMB open source community is jointly supported and initiated by the Natural Language Processing Laboratory of Tsinghua University and the Language Large Model Acceleration Technology Innovation Center of Zhiyuan Research Institute. The initiating team has a profound research foundation in natural language processing and pre-training models. In recent years, dozens of high-level papers have been published in top international conferences on model pre-training, prompt fine-tuning, and model compression technology.