SAM3 Docs update (#22932)

This commit is contained in:
Glenn Jocher 2025-12-12 15:16:35 +01:00 committed by GitHub
parent e0764aa55d
commit da2e575350
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 24 additions and 18 deletions

View File

@ -111,6 +111,16 @@ Explore the Ultralytics Docs, a comprehensive resource designed to help you unde
[:octicons-arrow-right-24: YOLO11 Models 🚀](models/yolo11.md)
- :material-select-all:{ .lg .middle }   **SAM 3: Segment Anything with Concepts 🚀 NEW**
***
Meta's latest SAM 3 with Promptable Concept Segmentation - segment all instances using text or image exemplars
***
[:octicons-arrow-right-24: SAM 3 Models](models/sam-3.md)
- :material-scale-balance:{ .lg .middle }   **Open Source, AGPL-3.0**
***

View File

@ -1,7 +1,7 @@
---
comments: true
description: Discover a variety of models supported by Ultralytics, including YOLOv3 to YOLO11, NAS, SAM, and RT-DETR for detection, segmentation, and more.
keywords: Ultralytics, supported models, YOLOv3, YOLOv4, YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, YOLOv10, YOLO11, SAM, SAM2, MobileSAM, FastSAM, YOLO-NAS, RT-DETR, YOLO-World, object detection, image segmentation, classification, pose estimation, multi-object tracking
keywords: Ultralytics, supported models, YOLOv3, YOLOv4, YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, YOLOv10, YOLO11, SAM, SAM2, SAM3, MobileSAM, FastSAM, YOLO-NAS, RT-DETR, YOLO-World, object detection, image segmentation, classification, pose estimation, multi-object tracking
---
# Models Supported by Ultralytics
@ -26,7 +26,7 @@ Here are some of the key models supported:
10. **[YOLO26](yolo26.md) ⚠️ Coming Soon**: Ultralytics' next-generation YOLO model optimized for edge deployment with end-to-end NMS-free inference.
11. **[Segment Anything Model (SAM)](sam.md)**: Meta's original Segment Anything Model (SAM).
12. **[Segment Anything Model 2 (SAM2)](sam-2.md)**: The next generation of Meta's Segment Anything Model for videos and images.
13. **[Segment Anything Model 3 (SAM3)](sam-3.md) ⚠️ Coming Soon**: Meta's upcoming third generation Segment Anything Model with enhanced memory and performance.
13. **[Segment Anything Model 3 (SAM3)](sam-3.md) 🚀 NEW**: Meta's third generation Segment Anything Model with Promptable Concept Segmentation for text and image exemplar-based segmentation.
14. **[Mobile Segment Anything Model (MobileSAM)](mobile-sam.md)**: MobileSAM for mobile applications, by Kyung Hee University.
15. **[Fast Segment Anything Model (FastSAM)](fast-sam.md)**: FastSAM by Image & Video Analysis Group, Institute of Automation, Chinese Academy of Sciences.
16. **[YOLO-NAS](yolo-nas.md)**: YOLO [Neural Architecture Search](https://www.ultralytics.com/glossary/neural-architecture-search-nas) (NAS) Models.

View File

@ -8,7 +8,7 @@ keywords: SAM 2, SAM 2.1, Segment Anything, video segmentation, image segmentati
!!! note "SAM Evolution"
SAM 2 builds upon the original [SAM](sam.md) with video segmentation capabilities. For the upcoming next-generation model with enhanced memory and 2× faster processing, see [SAM 3](sam-3.md).
SAM 2 builds upon the original [SAM](sam.md) with video segmentation capabilities. For Promptable Concept Segmentation with text and image exemplar prompts, see [SAM 3](sam-3.md).
<a href="https://colab.research.google.com/github/ultralytics/notebooks/blob/main/notebooks/inference-with-meta-sam-and-sam2-using-ultralytics-python-package.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Inference with Segment Anything 2 In Colab"></a>

View File

@ -6,15 +6,15 @@ keywords: SAM 3, Segment Anything 3, SAM3, SAM-3, video segmentation, image segm
# SAM 3: Segment Anything with Concepts
!!! note "Released — Ultralytics integration in progress"
!!! success "Now Available in Ultralytics"
Meta released SAM-3 on **November 20th, 2025**. Ultralytics is integrating the models now and will ship a package update with native support soon. In the meantime, you can follow the official SAM 3 README steps below to try the upstream release.
SAM 3 is fully integrated into the Ultralytics package as of **version 8.3.237** ([PR #22897](https://github.com/ultralytics/ultralytics/pull/22897)). Install or upgrade with `pip install -U ultralytics` to access all SAM 3 features including text-based concept segmentation, image exemplar prompts, and video tracking.
![SAM 3 Overview](https://github.com/ultralytics/docs/releases/download/0/sam-3-overview.webp)
**SAM 3** (Segment Anything Model 3) is Meta's released foundation model for **Promptable Concept Segmentation (PCS)**. Building upon [SAM 2](sam-2.md), SAM 3 introduces a fundamentally new capability: detecting, segmenting, and tracking **all instances** of a visual concept specified by text prompts, image exemplars, or both. Unlike previous SAM versions that segment single objects per prompt, SAM 3 can find and segment every occurrence of a concept appearing anywhere in images or videos, aligning with open-vocabulary goals in modern [instance segmentation](https://www.ultralytics.com/glossary/instance-segmentation).
Ultralytics is actively integrating SAM-3 into the `ultralytics` package. Until that release lands, you can experiment with the upstream Meta implementation using the official installation and usage steps below.
SAM 3 is now fully integrated into the `ultralytics` package, providing native support for concept segmentation with text prompts, image exemplar prompts, and video tracking capabilities.
## Overview
@ -108,10 +108,10 @@ SAM 3's scalable human- and model-in-the-loop data engine achieves **2× annotat
## Installation
SAM 3 will be available directly in the Ultralytics package once integration lands. Installation will remain:
SAM 3 is available in Ultralytics **version 8.3.237** and later. Install or upgrade with:
```bash
pip install ultralytics
pip install -U ultralytics
```
!!! warning "SAM 3 Model Weights Required"
@ -535,17 +535,13 @@ While SAM 3 represents a major advancement, it has certain limitations:
## FAQ
### When Will SAM 3 Be Released?
### When Was SAM 3 Released?
SAM 3 was released by Meta on **November 20th, 2025**. Ultralytics support is in progress and will ship in an upcoming package update with full docs for [predict mode](../modes/predict.md) and [track mode](../modes/track.md).
SAM 3 was released by Meta on **November 20th, 2025** and is fully integrated into Ultralytics as of **version 8.3.237** ([PR #22897](https://github.com/ultralytics/ultralytics/pull/22897)). Full support is available for [predict mode](../modes/predict.md) and [track mode](../modes/track.md).
### Will SAM 3 Be Integrated Into Ultralytics?
### Is SAM 3 Integrated Into Ultralytics?
Yes. SAM 3 will be supported in the Ultralytics Python package upon release, including concept segmentation, SAM 2–style visual prompts, and multi-object video tracking. You will be able to [export](../modes/export.md) to formats like [ONNX](../integrations/onnx.md) and [TensorRT](../integrations/tensorrt.md) for deployment, with streamlined [Python](../usage/python.md) and [CLI](../usage/cli.md) workflows.
!!! warning "Implementation Timeline"
Code examples in this documentation are preview versions showing intended usage patterns. Actual implementation will be available after Ultralytics completes integration.
Yes! SAM 3 is fully integrated into the Ultralytics Python package, including concept segmentation, SAM 2–style visual prompts, and multi-object video tracking. You can [export](../modes/export.md) to formats like [ONNX](../integrations/onnx.md) and [TensorRT](../integrations/tensorrt.md) for deployment, with streamlined [Python](../usage/python.md) and [CLI](../usage/cli.md) workflows.
### What Is Promptable Concept Segmentation (PCS)?

View File

@ -8,7 +8,7 @@ keywords: Segment Anything, SAM, image segmentation, promptable segmentation, ze
!!! note "SAM Evolution"
This is the original SAM model from Meta. For improved capabilities, see [SAM 2](sam-2.md) for video segmentation or [SAM 3](sam-3.md) for the upcoming next-generation model with enhanced memory and performance.
This is the original SAM model from Meta. For improved capabilities, see [SAM 2](sam-2.md) for video segmentation or [SAM 3](sam-3.md) for Promptable Concept Segmentation with text and image exemplar prompts.
<a href="https://colab.research.google.com/github/ultralytics/notebooks/blob/main/notebooks/inference-with-meta-sam-and-sam2-using-ultralytics-python-package.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="How to use Segment Anything In Colab"></a>

View File

@ -314,7 +314,7 @@ nav:
- YOLO26 🚀 Coming soon: models/yolo26.md
- SAM (Segment Anything Model): models/sam.md
- SAM 2 (Segment Anything Model 2): models/sam-2.md
- SAM 3 (Segment Anything Model 3) 🚀 Coming soon: models/sam-3.md
- SAM 3 (Segment Anything Model 3): models/sam-3.md
- MobileSAM (Mobile Segment Anything Model): models/mobile-sam.md
- FastSAM (Fast Segment Anything Model): models/fast-sam.md
- YOLO-NAS (Neural Architecture Search): models/yolo-nas.md