Elgato_dark/ultralytics
1
0
Fork 0
mirror of https://github.com/ultralytics/ultralytics synced 2026-04-21 14:07:18 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 8

Add YOLO26 Kaggle models (#23289)

Browse source
This commit is contained in:
Glenn Jocher 2026-01-16 19:20:23 +00:00 committed by GitHub
parent d692ec7d37
commit fb784bef12
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
11 changed files with 26 additions and 27 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
README.md
View file

@ -15,7 +15,7 @@
<br>
<a href="https://console.paperspace.com/github/ultralytics/ultralytics"><img src="https://assets.paperspace.io/img/gradient-badge.svg" alt="Run Ultralytics on Gradient"></a>
<a href="https://colab.research.google.com/github/ultralytics/ultralytics/blob/main/examples/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open Ultralytics In Colab"></a>
<a href="https://www.kaggle.com/models/ultralytics/yolo11"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open Ultralytics In Kaggle"></a>
<a href="https://www.kaggle.com/models/ultralytics/yolo26"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open Ultralytics In Kaggle"></a>
<a href="https://mybinder.org/v2/gh/ultralytics/ultralytics/HEAD?labpath=examples%2Ftutorial.ipynb"><img src="https://mybinder.org/badge_logo.svg" alt="Open Ultralytics In Binder"></a>
</div>
</div>

2
README.zh-CN.md
View file

@ -15,7 +15,7 @@
<br>
<a href="https://console.paperspace.com/github/ultralytics/ultralytics"><img src="https://assets.paperspace.io/img/gradient-badge.svg" alt="Run Ultralytics on Gradient"></a>
<a href="https://colab.research.google.com/github/ultralytics/ultralytics/blob/main/examples/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open Ultralytics In Colab"></a>
<a href="https://www.kaggle.com/models/ultralytics/yolo11"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open Ultralytics In Kaggle"></a>
<a href="https://www.kaggle.com/models/ultralytics/yolo26"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open Ultralytics In Kaggle"></a>
<a href="https://mybinder.org/v2/gh/ultralytics/ultralytics/HEAD?labpath=examples%2Ftutorial.ipynb"><img src="https://mybinder.org/badge_logo.svg" alt="Open Ultralytics In Binder"></a>
</div>
</div>

2
docs/en/index.md
View file

@ -29,7 +29,7 @@ keywords: Ultralytics, YOLO, YOLO26, YOLO11, object detection, image segmentatio
<br>
<a href="https://console.paperspace.com/github/ultralytics/ultralytics"><img src="https://assets.paperspace.io/img/gradient-badge.svg" alt="Run Ultralytics on Gradient"></a>
<a href="https://colab.research.google.com/github/ultralytics/ultralytics/blob/main/examples/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open Ultralytics In Colab"></a>
<a href="https://www.kaggle.com/models/ultralytics/yolo11"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open Ultralytics In Kaggle"></a>
<a href="https://www.kaggle.com/models/ultralytics/yolo26"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open Ultralytics In Kaggle"></a>
<a href="https://mybinder.org/v2/gh/ultralytics/ultralytics/HEAD?labpath=examples%2Ftutorial.ipynb"><img src="https://mybinder.org/badge_logo.svg" alt="Open Ultralytics In Binder"></a>
<br><br>
</div>

2
docs/en/platform/data/annotation.md
View file

@ -170,7 +170,7 @@ Use trained YOLO models to automatically label images:
You can use:
- Official Ultralytics models (YOLO11n, YOLO11s, etc.)
- Official Ultralytics models (YOLO26n, YOLO26s, etc.)
- Your own trained models from the Platform
## Class Management

6
docs/en/platform/deploy/endpoints.md
View file

@ -276,9 +276,9 @@ Cold start varies by model size:
| Model | Cold Start |
| ------- | ---------- |
| YOLO11n | ~2 seconds |
| YOLO11m | ~3 seconds |
| YOLO11x | ~5 seconds |
| YOLO26n | ~2 seconds |
| YOLO26m | ~3 seconds |
| YOLO26x | ~5 seconds |
Set min instances > 0 to eliminate cold starts.

2
docs/en/platform/index.md
View file

@ -48,7 +48,7 @@ Ultralytics Platform is designed to replace fragmented ML tooling with a unified
- **HuggingFace** - Model deployment
- **Arize** - Monitoring
All in one platform with native support for YOLO11 and YOLO26 models.
All in one platform with native support for YOLO26 and YOLO11 models.
## Workflow: Data → Train → Deploy

12
docs/en/platform/quickstart.md
View file

@ -135,17 +135,17 @@ From your project, click "Train Model" to start cloud training.
### Training Configuration
1. **Select Dataset**: Choose from your uploaded datasets
2. **Choose Model**: Select a base model (YOLO11n, YOLO11s, etc.)
2. **Choose Model**: Select a base model (YOLO26n, YOLO26s, etc.)
3. **Set Epochs**: Number of training iterations
4. **Select GPU**: Choose compute resources
| Model | Size | Speed | Accuracy |
| ------- | ----------- | -------- | -------- |
| YOLO11n | Nano | Fastest | Good |
| YOLO11s | Small | Fast | Better |
| YOLO11m | Medium | Moderate | High |
| YOLO11l | Large | Slower | Higher |
| YOLO11x | Extra Large | Slowest | Best |
| YOLO26n | Nano | Fastest | Good |
| YOLO26s | Small | Fast | Better |
| YOLO26m | Medium | Moderate | High |
| YOLO26l | Large | Slower | Higher |
| YOLO26x | Extra Large | Slowest | Best |
### Monitor Training

18
docs/en/platform/train/cloud-training.md
View file

@ -44,7 +44,7 @@ Select base model and parameters:
| Parameter | Description | Default |
| -------------- | --------------------------------------- | ------- |
| **Model** | Base architecture (YOLO11n, s, m, l, x) | YOLO11n |
| **Model** | Base architecture (YOLO26n, s, m, l, x) | YOLO26n |
| **Epochs** | Number of training iterations | 100 |
| **Image Size** | Input resolution | 640 |
| **Batch Size** | Samples per iteration | Auto |
@ -242,11 +242,11 @@ After training, view detailed costs in the **Billing** tab:
| Model | Parameters | Best For |
| ------- | ---------- | ----------------------- |
| YOLO11n | 2.6M | Real-time, edge devices |
| YOLO11s | 9.4M | Balanced speed/accuracy |
| YOLO11m | 20.1M | Higher accuracy |
| YOLO11l | 25.3M | Production accuracy |
| YOLO11x | 56.9M | Maximum accuracy |
| YOLO26n | 2.4M | Real-time, edge devices |
| YOLO26s | 9.5M | Balanced speed/accuracy |
| YOLO26m | 20.4M | Higher accuracy |
| YOLO26l | 24.8M | Production accuracy |
| YOLO26x | 55.7M | Maximum accuracy |
### Optimize Training Time
@ -279,9 +279,9 @@ Typical times (1000 images, 100 epochs):
| Model | RTX 4090 | A100 |
| ------- | -------- | ------ |
| YOLO11n | 30 min | 20 min |
| YOLO11m | 60 min | 40 min |
| YOLO11x | 120 min | 80 min |
| YOLO26n | 30 min | 20 min |
| YOLO26m | 60 min | 40 min |
| YOLO26x | 120 min | 80 min |
### Can I train overnight?

2
docs/en/platform/train/index.md
View file

@ -96,7 +96,7 @@ Training time depends on:
- Number of epochs
- GPU type selected
A typical training run with 1000 images, YOLO11n, 100 epochs on RTX 4090 takes about 30-60 minutes.
A typical training run with 1000 images, YOLO26n, 100 epochs on RTX 4090 takes about 30-60 minutes.
### Can I train multiple models simultaneously?

4
docs/en/platform/train/models.md
View file

@ -31,7 +31,7 @@ Supported model formats:
After upload, the Platform parses model metadata:
- Task type (detect, segment, pose, OBB, classify)
- Architecture (YOLO11n, YOLO11s, etc.)
- Architecture (YOLO26n, YOLO26s, etc.)
- Class names and count
- Input size and parameters
@ -192,8 +192,8 @@ Remove a model you no longer need:
Ultralytics Platform supports all YOLO architectures:
- **YOLO26**: n, s, m, l, x variants (recommended)
- **YOLO11**: n, s, m, l, x variants
- **YOLO26**: Latest generation (when available)
- **YOLOv10**: Legacy support
- **YOLOv8**: Legacy support
- **YOLOv5**: Legacy support

1
pyproject.toml
View file

@ -111,7 +111,6 @@ logging = [
"mlflow", # https://docs.ultralytics.com/integrations/mlflow/
]
extra = [
"hub-sdk>=0.0.12", # Ultralytics HUB
"ipython", # interactive notebook
"albumentations>=1.4.6", # training augmentations
"faster-coco-eval>=1.6.7", # COCO mAP