Video-Text-to-Text
Transformers
Safetensors
English
Keye
image-feature-extraction
multimodal
custom_code
Instructions to use FriendliAI/Keye-VL-8B-Preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FriendliAI/Keye-VL-8B-Preview with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("FriendliAI/Keye-VL-8B-Preview", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "min_pixels": 102400, | |
| "max_pixels": 1003520, | |
| "patch_size": 14, | |
| "temporal_patch_size": 1, | |
| "merge_size": 2, | |
| "image_mean": [ | |
| 0.5,0.5,0.5 | |
| ], | |
| "image_std": [ | |
| 0.5,0.5,0.5 | |
| ], | |
| "processor_class": "KeyeProcessor", | |
| "auto_map": { | |
| "AutoProcessor": "processing_keye.KeyeProcessor", | |
| "AutoImageProcessor": "image_processing_keye.SiglipImageProcessor" | |
| } | |
| } | |