InternVL2_5-78B-MPO

InternVL2.5-MPO is a series of multi-modal large-scale language models based on InternVL2.5 and Mixed Preference Optimization (MPO). It performs well on multi-modal tasks by integrating the newly incrementally pretrained InternViT with multiple pretrained large language models (LLMs), such as InternLM 2.5 and Qwen 2.5, using randomly initialized MLP projectors. This model series was trained on the multi-modal reasoning preference data set MMPR, which contains approximately 3 million samples. Through effective data construction processes and hybrid preference optimization technology, the model's reasoning capabilities and answer quality are improved.

The target audience includes researchers, developers and enterprises, and is suitable for scenarios that require multi-modal understanding and generation, such as smart assistants, content creation, image and video analysis, etc. The model's high performance and flexibility make it ideal for handling complex multi-modal tasks.