研究报告
Abstract We report the development of GPT-4, a large-scale, multimodal model which can accept image and text inputs and produce text outputs.
暂无关联机构
← 返回知识图谱