metadata
license: apache-2.0
language:
- en
base_model:
- OpenGVLab/InternVideo2_distillation_models
pipeline_tag: video-classification
InternVideo2-B14
Cross-Modal and Vision-Language Model Checkpoints
This repository hosts pre-trained model checkpoints for cross-modal video-text understanding, vision-language alignment, and efficient deployment. Below is a summary of included files:
| Filename | Size | Description |
|---|---|---|
| cross_mamba_film_warmup.pt | 504 MB | A cross-modal checkpoint combining vision and text using FiLM (Feature-wise Linear Modulation) and Mamba layers |
| internvideo2_clip.pt | 5.55 MB | CLIP component of InternVideo2-B14 |
| internvideo2_vision.pt | 205 MB | Vision encoder backbone for InternVideo2-B14 |
| mobileclip_blt.pt | 599 MB | Lightweight MobileCLIP variant (BLT) |