Models

Ultralytics Platform provides comprehensive model management for training, analyzing, and deploying YOLO models. Upload pretrained models or train new ones directly on the Platform.

Upload Model

Upload existing model weights to the Platform:

1. Navigate to your project
2. Click Upload Model
3. Select your .pt file
4. Add name and description
5. Click Upload

Supported model formats:

After upload, the Platform parses model metadata:

• Task type (detect, segment, pose, OBB, classify)
• Architecture (YOLO26n, YOLO26s, etc.)
• Class names and count
• Input size and parameters

Train Model

Train a new model directly on the Platform:

1. Navigate to your project
2. Click Train Model
3. Select dataset
4. Choose base model
5. Configure training parameters
6. Start training

See Cloud Training for detailed instructions.

Model Overview

Each model page displays:

Training Metrics

View real-time and historical training metrics:

Loss Curves

Performance Metrics

Validation Plots

After training completes, view detailed validation analysis:

Confusion Matrix

Interactive heatmap showing prediction accuracy per class:

PR/F1 Curves

Performance curves at different confidence thresholds:

Export Model

Export your model to 17 deployment formats:

1. Navigate to the Export tab
2. Select target format
3. Click Export
4. Download when complete

Supported Formats (17 total)

Format Selection Guide

For NVIDIA GPUs: Use TensorRT for maximum speed

For Intel Hardware: Use OpenVINO for Intel CPUs, GPUs, and VPUs

For Apple Devices: Use CoreML for iOS, macOS, Apple Silicon

For Android: Use TF Lite or NCNN for best performance

For Web Browsers: Use TF.js or ONNX (with ONNX Runtime Web)

For Edge Devices: Use TF Edge TPU for Coral, RKNN for Rockchip

For General Compatibility: Use ONNX — works with most inference runtimes

Dataset Linking

Models can be linked to their source dataset:

• View which dataset was used for training
• Access dataset from model page
• Track data lineage

When training with Platform datasets using the ul:// URI format, linking is automatic.

Visibility Settings

Control who can see your model:

To change visibility:

1. Open model actions menu
2. Click Edit
3. Toggle visibility
4. Click Save

Delete Model

Remove a model you no longer need:

1. Open model actions menu
2. Click Delete
3. Confirm deletion

FAQ

What model architectures are supported?

Ultralytics Platform supports all YOLO architectures:

• YOLO26: n, s, m, l, x variants (recommended)
• YOLO11: n, s, m, l, x variants
• YOLOv10: Legacy support
• YOLOv8: Legacy support
• YOLOv5: Legacy support

Can I download my trained model?

Yes, download your model weights from the model page:

1. Click the download icon
2. Select format (original .pt or exported)
3. Download starts automatically

How do I compare models across projects?

Currently, model comparison is within projects. To compare across projects:

1. Transfer models to a single project, or
2. Export metrics and compare externally

What's the maximum model size?

There's no strict limit, but very large models (>2GB) may have longer upload and processing times.

Can I fine-tune pretrained models?

Yes! Upload a pretrained model, then start training from that checkpoint with your dataset. The Platform automatically uses the uploaded model as the starting point.

📅 Created 20 days ago ✏️ Updated 13 days ago

Comments