Authors: InternLM-XComposer Team

Affiliation: Shanghai Artificial Intelligence Laboratory

Description: InternLM-XComposer2-4KHD: Scaling the Resolution of Large Vision-Language Models up to 4KHD and Beyond

1. Generalist model with 4H Resolution understanding.
2. End-to-end model, no OCR pipeline.
3. No specialist fine-tuning.
see https://github.com/InternLM/InternLM-XComposer for more details.