|
--- |
|
tags: |
|
- text-to-image |
|
- lora |
|
- diffusers |
|
- template:diffusion-lora |
|
widget: |
|
- text: 'CMS 3D Art, Captured at eye-level on a vibrant purple backdrop, a medium-sized figurine of a cartoon character, dressed in a blue sweater, black shorts, and black and white sneakers. The figurine is holding a gray cup with a white straw in it, and a green bag with yellow smiley faces on it. The characters outfit is adorned with a black strap, a white tag, and green socks with white stripes. The figures eyes are blue, and he has a white hat on his head. The background is a soft lavender shade, adding a pop of color to the scene.' |
|
output: |
|
url: images/1.png |
|
- text: 'CMS 3D Art, a medium-sized cartoon figure stands on a vibrant green backdrop. The figure is dressed in a pink button-down shirt, adorned with a vibrant orange and yellow floral pattern, and a pair of gray sweatpants. His shoes are adorned with blue and yellow sneakers, adding a pop of color to the scene. He is holding a white cup with a black straw in it, while his right hand is draped over his right shoulder. His left hand is resting on a large green object, which is positioned in front of him. The background is a vibrant shade of green, creating a striking contrast to the figure.' |
|
output: |
|
url: images/22.png |
|
- text: 'CMS 3D Art, Captured from a low-angle perspective on a vibrant blue backdrop, a figurine stands on a blue carpet. The figurine, dressed in a black coat, adorned with green accents, is adorned with a white emblem on the chest, adding a pop of color to the scene. The figure, posed in a striking pose, is holding a black bag in one hand, while the other hand rests on the bag in the other. Its legs are spread out in a way that creates a striking contrast to the blue carpet, creating a stark contrast.' |
|
output: |
|
url: images/33.png |
|
base_model: black-forest-labs/FLUX.1-dev |
|
instance_prompt: CMS 3D Art |
|
license: apache-2.0 |
|
--- |
|
 |
|
|
|
<Gallery /> |
|
# Model description for CMS 3D Art |
|
|
|
Image Processing Parameters |
|
|
|
| Parameter | Value | Parameter | Value | |
|
|---------------------------|--------|---------------------------|--------| |
|
| LR Scheduler | constant | Noise Offset | 0.03 | |
|
| Optimizer | AdamW | Multires Noise Discount | 0.1 | |
|
| Network Dim | 64 | Multires Noise Iterations | 10 | |
|
| Network Alpha | 32 | Repeat & Steps | 26 & 3250 | |
|
| Epoch | 25 | Save Every N Epochs | 1 | |
|
|
|
Labeling: florence2-en(natural language & English) |
|
|
|
Total Images Used for Training : 34 |
|
|
|
## Best Dimensions & Inference |
|
|
|
| **Dimensions** | **Aspect Ratio** | **Recommendation** | |
|
|-----------------|------------------|---------------------------| |
|
| 1280 x 832 | 3:2 | Best | |
|
| 1024 x 1024 | 1:1 | Default | |
|
|
|
### Inference Range |
|
|
|
- **Recommended Inference Steps:** 30–35 |
|
|
|
## Setting Up |
|
```python |
|
import torch |
|
from pipelines import DiffusionPipeline |
|
|
|
base_model = "black-forest-labs/FLUX.1-dev" |
|
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16) |
|
|
|
lora_repo = "strangerzonehf/CMS-3D-Art" |
|
trigger_word = "CMS 3D Art" |
|
pipe.load_lora_weights(lora_repo) |
|
|
|
device = torch.device("cuda") |
|
pipe.to(device) |
|
``` |
|
|
|
## Trigger words |
|
|
|
You should use `CMS 3D Art` to trigger the image generation. |
|
|
|
|
|
## Download model |
|
|
|
Weights for this model are available in Safetensors format. |
|
|
|
[Download](/strangerzonehf/CMS-3D-Art/tree/main) them in the Files & versions tab. |
|
|
