OpenGVLab

vision-language-models | multimodal-evaluation

Overview

OpenGVLab is the organization behind InternVL, InternLM-V, and M3I-Eval benchmarks. It provides a suite of open-source vision-language models and evaluation tools.

Key Projects

• InternVL: Large-scale vision-language foundation model
• InternLM-V: Vision-language variant of InternLM
• M3I-Eval: Multi-modal multi-choice evaluation benchmark

Relationship to Other Projects

• Related to OpenGVLab (self-reference via the organization)
• Competes with OpenAI CLIP and LLaVA in the VLM space

References

• GitHub: https://github.com/opengvlab