Text-to-Image
Diffusers
Safetensors
diffusers-training
stable-diffusion-xl
stable-diffusion-xl-diffusers
Instructions to use mapo-t2i/mapo-beta with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use mapo-t2i/mapo-beta with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("mapo-t2i/mapo-beta", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Update README.md
Browse files
README.md
CHANGED
|
@@ -104,6 +104,36 @@ We evaluated this checkpoint in the Imgsys public benchmark. MaPO was able to ou
|
|
| 104 |
<img src="https://mapo-t2i.github.io/static/images/imgsys.png" width=750/>
|
| 105 |
</div>
|
| 106 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
|
| 108 |
## Inference
|
| 109 |
|
|
|
|
| 104 |
<img src="https://mapo-t2i.github.io/static/images/imgsys.png" width=750/>
|
| 105 |
</div>
|
| 106 |
|
| 107 |
+
The table below reports memory efficiency of MaPO, making it a better alternative for alignment fine-tuning of diffusion models:
|
| 108 |
+
|
| 109 |
+
<table>
|
| 110 |
+
<caption>Computational costs of Diffusion-DPO and MaPO</caption>
|
| 111 |
+
<thead>
|
| 112 |
+
<tr>
|
| 113 |
+
<th></th>
|
| 114 |
+
<th>Diffusion-DPO</th>
|
| 115 |
+
<th>MaPO <span class="ours">(Ours)</span></th>
|
| 116 |
+
</tr>
|
| 117 |
+
</thead>
|
| 118 |
+
<tbody>
|
| 119 |
+
<tr>
|
| 120 |
+
<td>Time (↓)</td>
|
| 121 |
+
<td>63.5</td>
|
| 122 |
+
<td>54.3 (-14.5%)</td>
|
| 123 |
+
</tr>
|
| 124 |
+
<tr>
|
| 125 |
+
<td>GPU Mem. (↓)</td>
|
| 126 |
+
<td>55.9</td>
|
| 127 |
+
<td>46.1 (-17.5%)</td>
|
| 128 |
+
</tr>
|
| 129 |
+
<tr>
|
| 130 |
+
<td>Max Batch (↑)</td>
|
| 131 |
+
<td>4</td>
|
| 132 |
+
<td>16 (×4)</td>
|
| 133 |
+
</tr>
|
| 134 |
+
</tbody>
|
| 135 |
+
</table>
|
| 136 |
+
|
| 137 |
|
| 138 |
## Inference
|
| 139 |
|