๐ Sana
Sana is a text - to - image framework that can efficiently generate high - resolution images up to 4096 ร 4096. It can synthesize high - quality images with strong text - image alignment at a fast speed and can be deployed on laptop GPUs.
๐ Quick Start
The source code of Sana is available at GitHub. You can explore the model through the provided links above.
โจ Features
- High - Resolution Image Generation: Capable of generating images up to 4096 ร 4096 resolution.
- Fast and High - Quality: Can synthesize high - quality images with strong text - image alignment at a remarkable speed, and is deployable on laptop GPUs.
- Multi - language Support: Supports English, Chinese, and Emoji, as well as all mixed prompts.
๐ Documentation
Compare with base model
Property |
Details |
Model Type |
Linear - Diffusion - Transformer - based text - to - image generative model |
Model Size |
1648M parameters |
Model Resolution |
Developed to generate 1024px based images with multi - scale height and width |
License |
NSCL v2 - custom. Governing Terms: NVIDIA License. Additional Information: [Gemma Terms of Use |
Model Description |
A model that can be used to generate and modify images based on text prompts. It is a Linear Diffusion Transformer that uses one fixed, pretrained text encoder ([Gemma2 - 2B - IT](https://huggingface.co/google/gemma - 2 - 2b - it)) and one 32x spatial - compressed latent feature encoder ([DC - AE](https://hanlab.mit.edu/projects/dc - ae)). |
Special |
Fine - tuned from the base model [Efficient - Large - Model/Sana_1600M_1024px](https://huggingface.co/Efficient - Large - Model/Sana_1600M_1024px) and supports Emoji, Chinese, English, and all mixed prompts. |
Resources for more information |
Check out our GitHub Repository and the Sana report on arXiv. |
Model |
Language |
Sana_1600M_1024px |
English |
Sana_1600M_1024px_MultiLing |
English, Chinese, Emoji |
Model |
Sample - 1 |
Sample - 2 |
Sample - 3 |
Sample - 4 |
Sana_1600M_1024px |
 |
 |
 |
 |
Sana_1600M_1024px_MultiLing |
 |
 |
 |
 |
Prompt |
๐ฏ ็ฉฟ็ ๐ ๅน ๐ท |
็ซ Wearing ๐ถ flying on the ๅฝฉ่น with ๐น in the โ๏ธ |
๐ฆ teaching ๐ฏ to catch ๐ฆ |
้่ฒ ๐
ไธ็้ฟๅ, traditional Chinese style |
Model Sources
For research purposes, we recommend our generative - models
Github repository (https://github.com/NVlabs/Sana), which is more suitable for both training and inference and for which most advanced diffusion sampler like Flow - DPM - Solver is integrated. [MIT Han - Lab](https://nv - sana.mit.edu/) provides free Sana inference.
- Repository: https://github.com/NVlabs/Sana
- Demo: https://nv - sana.mit.edu/
๐งจ Diffusers
PR developing: Sana and DC - AE
๐ก Usage
Direct Use
The model is intended for research purposes only. Possible research areas and tasks include:
- Generation of artworks and use in design and other artistic processes.
- Applications in educational or creative tools.
- Research on generative models.
- Safe deployment of models which have the potential to generate harmful content.
- Probing and understanding the limitations and biases of generative models.
Out - of - Scope Use
The model was not trained to be factual or true representations of people or events, and therefore using the model to generate such content is out - of - scope for the abilities of this model.
๐ง Technical Details
Limitations
- The model does not achieve perfect photorealism.
- The model cannot render complex legible text.
- Fingers, etc. in general may not be generated properly.
- The autoencoding part of the model is lossy.
Bias
While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.
๐ License
The model is under NSCL v2 - custom. Governing Terms: NVIDIA License. Additional Information: Gemma Terms of Use | Google AI for Developers for Gemma - 2 - 2B - IT, Gemma Prohibited Use Policy | Google AI for Developers.