# Training Resolutions by Model Type

Training resolution affects model accuracy, inference speed, and training time. Each model architecture has a default resolution that balances these factors. By default, Roboflow suggests the default training resolution for the selected model architecture.

The table below shows the default training resolution for each model architecture and size. You can override these defaults by configuring the resize preprocessing step when creating a new [Dataset Version](https://docs.roboflow.com/datasets/dataset-versions).

### Object Detection

<table><thead><tr><th>Model Type</th><th width="273.3359375">Family &#x26; Size</th><th>Default Training Resolution</th></tr></thead><tbody><tr><td>Object Detection</td><td>RF-DETR Nano</td><td>384×384</td></tr><tr><td>Object Detection</td><td>RF-DETR Small</td><td>512×512</td></tr><tr><td>Object Detection</td><td>RF-DETR Medium</td><td>576×576</td></tr><tr><td>Object Detection</td><td>RF-DETR Large</td><td>704×704</td></tr><tr><td>Object Detection</td><td>RF-DETR X Large</td><td>700x700</td></tr><tr><td>Object Detection</td><td>RF-DETR 2X Large</td><td>880x880</td></tr><tr><td>Object Detection</td><td>Roboflow 3.0 - Fast</td><td>640×640</td></tr><tr><td>Object Detection</td><td>Roboflow 3.0 - Accurate</td><td>640×640</td></tr><tr><td>Object Detection</td><td>Roboflow 3.0 - Medium</td><td>640×640</td></tr><tr><td>Object Detection</td><td>Roboflow 3.0 - Large</td><td>640×640</td></tr><tr><td>Object Detection</td><td>Roboflow 3.0 - Extra Large</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv26(n/s/m/l/x)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv12 (n/s/m/l/x)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv11 (n/s/m/l/x)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv10 (n/s/m/b/l/x)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv9 (s/m/c/e)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv8 (n/s/m/l/x)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv5 (n/s/m/l/x)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLOv7 (legacy)</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLO‑NAS Small</td><td>640×640</td></tr><tr><td>Object Detection</td><td>YOLO‑NAS Medium</td><td>640×640</td></tr><tr><td>Object Detection</td><td>Roboflow Instant</td><td>1008x1008</td></tr></tbody></table>

### Instance Segmentation

<table><thead><tr><th>Model Type</th><th width="272.8203125">Family &#x26; Size</th><th>Default Training Resolution</th></tr></thead><tbody><tr><td>Instance Segmentation</td><td>RF-DETR Seg Nano</td><td>312x312</td></tr><tr><td>Instance Segmentation</td><td>RF-DETR Seg Small</td><td>384x384</td></tr><tr><td>Instance Segmentation</td><td>RF-DETR Seg Medium</td><td>432x432</td></tr><tr><td>Instance Segmentation</td><td>RF-DETR Seg Large</td><td>504x504</td></tr><tr><td>Instance Segmentation</td><td>RF-DETR Seg X Large</td><td>624x624</td></tr><tr><td>Instance Segmentation</td><td>RF-DETR Seg 2X Large</td><td>768x768</td></tr><tr><td>Instance Segmentation</td><td>Roboflow 3.0 - Fast (Seg)</td><td>640×640</td></tr><tr><td>Instance Segmentation</td><td>Roboflow 3.0 - Accurate (Seg)</td><td>640×640</td></tr><tr><td>Instance Segmentation</td><td>Roboflow 3.0 - Medium (Seg)</td><td>640×640</td></tr><tr><td>Instance Segmentation</td><td>Roboflow 3.0 - Large (Seg)</td><td>640×640</td></tr><tr><td>Instance Segmentation</td><td>Roboflow 3.0 - Extra Large (Seg)</td><td>640×640</td></tr><tr><td>Instance Segmentation</td><td>YOLO-seg (v8/10/11/12)</td><td>640×640</td></tr><tr><td>Instance Segmentation</td><td>SAM3 (Segment Anything 3)</td><td>1008x1008</td></tr><tr><td>Instance Segmentation</td><td>Semantic segmentation (DeepLabV3+)</td><td>≥ 512×512</td></tr></tbody></table>

### Classification & Pose

<table><thead><tr><th>Model Type</th><th width="272.66796875">Family &#x26; Size</th><th>Default Training Resolution</th></tr></thead><tbody><tr><td>Classification &#x26; Pose</td><td>Resnet-18/34/50</td><td>224x224</td></tr><tr><td>Classification &#x26; Pose</td><td>YOLO-cls (v8/11)</td><td>224x224</td></tr><tr><td>Classification &#x26; Pose</td><td>Vision Transformer (ViT)</td><td>224x224</td></tr><tr><td>Classification &#x26; Pose</td><td>YOLO-pose (keypoints)</td><td>640x640</td></tr></tbody></table>

### Multimodal/VLM

<table><thead><tr><th>Model Type</th><th width="272.96484375">Family &#x26; Size</th><th>Default Training Resolution</th></tr></thead><tbody><tr><td>Multimodal/VLM</td><td>PaliGemma 2 - 3 B</td><td>448x448</td></tr><tr><td>Multimodal/VLM</td><td>PaliGemma 2 - 10 B/28 B</td><td>448x448</td></tr><tr><td>Multimodal/VLM</td><td>Florence-2</td><td>448x448</td></tr><tr><td>Multimodal/VLM</td><td>QWEN 2.5 VL</td><td>448x448</td></tr><tr><td>Multimodal/VLM</td><td>SmolVLM2</td><td>384x384</td></tr></tbody></table>