| | --- |
| | license: other |
| | license_name: sam-license |
| | license_link: LICENSE |
| | tags: |
| | - segmentation |
| | - image-segmentation |
| | - video-segmentation |
| | - sam3 |
| | - meta |
| | - pytorch |
| | - safetensors |
| | pipeline_tag: image-segmentation |
| | base_model: facebook/sam3 |
| | --- |
| | |
| | # SAM3 — Segment Anything Model 3 |
| |
|
| | This is an **ungated mirror** of Meta's [SAM3](https://github.com/facebookresearch/sam3) model weights, redistributed under the [SAM License](LICENSE) for easier access. |
| |
|
| | ## What is SAM3? |
| |
|
| | SAM3 (Segment Anything Model 3) is Meta AI's state-of-the-art foundation model for **promptable segmentation** in images and videos. It supports: |
| |
|
| | - **Text prompts** — segment objects by describing them (e.g. *"the dog"*) |
| | - **Bounding box prompts** — precise region targeting |
| | - **Point prompts** — click-to-segment |
| | - **Video tracking** — track and segment objects across video frames |
| | - **Automatic mask generation** — segment all objects in a scene |
| |
|
| | ## Files |
| |
|
| | | File | Description | |
| | |---|---| |
| | | `sam3.safetensors` | Model weights (safetensors format, converted from official `.pt`) | |
| | | `LICENSE` | SAM License (required for redistribution) | |
| |
|
| | ## ⚠️ Conversion Note |
| |
|
| | The `sam3.safetensors` in this repo is converted **directly from the official `sam3.pt`** checkpoint using: |
| |
|
| | ```python |
| | import torch |
| | from safetensors.torch import save_file |
| | |
| | state_dict = torch.load("sam3.pt", map_location="cpu") |
| | save_file(state_dict, "sam3.safetensors") |
| | ``` |
| |
|
| | This preserves the original key names (`detector.*`, `tracker.*`). Do **not** use the HuggingFace Transformers-format safetensors from `facebook/sam3` — those have different key names (`detector_model.*`) and are **incompatible** with `build_sam3_video_model()`. |
| |
|
| | ## Usage |
| |
|
| | ```python |
| | # With the official sam3 package |
| | pip install sam3 |
| | |
| | from sam3.model_builder import build_sam3_image_model |
| | model = build_sam3_image_model(checkpoint_path="sam3.safetensors") |
| | ``` |
| |
|
| | ## License |
| |
|
| | This model is distributed under the **SAM License** — see the [LICENSE](LICENSE) file. Key points: |
| |
|
| | - ✅ Commercial use permitted |
| | - ✅ Redistribution permitted (with license included) |
| | - ✅ Derivative works permitted |
| | - ❌ No military/warfare, nuclear, or espionage use |
| | - ❌ No reverse engineering |
| |
|
| | ## Credits |
| |
|
| | - **Original model by**: [Meta AI (FAIR)](https://github.com/facebookresearch/sam3) |
| | - **Original HuggingFace repo**: [facebook/sam3](https://huggingface.co/facebook/sam3) |
| | - **Paper**: *Segment Anything Model 3* (submitted to ICLR 2026) |
| | - **Redistributed by**: [Æmotion Studio](https://huggingface.co/AEmotionStudio) for use with [ComfyUI-FFMPEGA](https://github.com/AEmotionStudio/ComfyUI-FFMPEGA) |
| |
|
