Mask Generation
Transformers
Safetensors
edgetam_video
feature-extraction
libreyolo
edgetam
promptable-segmentation
image-segmentation
Instructions to use LibreYOLO/LibreEdgeTAM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LibreYOLO/LibreEdgeTAM with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("mask-generation", model="LibreYOLO/LibreEdgeTAM")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("LibreYOLO/LibreEdgeTAM") model = AutoModel.from_pretrained("LibreYOLO/LibreEdgeTAM") - Notebooks
- Google Colab
- Kaggle
| LibreEdgeTAM provenance notice | |
| Official model | |
| -------------- | |
| Repository: https://huggingface.co/facebook/EdgeTAM | |
| Revision: 14d7ecc48c656b94e5184519f698cd5386c5a2bf | |
| File: edgetam.pt | |
| SHA-256: ed2d4850b8792c239689b043c47046ec239b6e808a3d9b6ae676c803fd8780df | |
| Official source and license | |
| --------------------------- | |
| Repository: https://github.com/facebookresearch/EdgeTAM | |
| Commit: 7711e012a30a2402c4eaab637bdb00a521302c91 | |
| License: Apache License 2.0 | |
| LICENSE SHA-256: c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4 | |
| Conversion implementation | |
| ------------------------- | |
| Repository: https://github.com/huggingface/transformers | |
| Commit: bd37c453544e83eb875ed3608980a1660376007a | |
| File: src/transformers/models/edgetam_video/convert_edgetam_video_to_hf.py | |
| Copyright 2025 The HuggingFace Inc. team | |
| License: Apache License 2.0 | |
| Hash-pinned snapshot metadata source and conversion reference | |
| ------------------------------------------------------------- | |
| Repository: https://huggingface.co/yonigozlan/EdgeTAM-hf | |
| Revision: c266ce53b3fc00f0f495b583f6a116c4e57f53bb | |
| License declared by reference model card: Apache License 2.0 | |
| model.safetensors SHA-256: 8858f8e4757b0b96dab8763f296ecffd845efbbbf698f64163cfa20a63d5fff4 | |
| Files copied byte-for-byte after SHA-256 verification: | |
| .gitattributes | |
| config.json | |
| preprocessor_config.json | |
| processor_config.json | |
| video_preprocessor_config.json | |
| LibreYOLO modifications | |
| ----------------------- | |
| The official checkpoint weights were independently safe-loaded and converted | |
| through lossless key remapping, key/value tensor splitting, and point-embedding | |
| concatenation; model.safetensors was not copied from the reference. The result | |
| was strict-loaded into EdgeTamVideoModel. All 984 state tensors were | |
| checked for exact equality with the pinned reference snapshot. The five | |
| non-weight files listed above were copied from that reference revision. The raw | |
| .pt checkpoint was omitted from this mirror. | |
| Generated model.safetensors SHA-256: 8858f8e4757b0b96dab8763f296ecffd845efbbbf698f64163cfa20a63d5fff4 | |