Efficient-Large-Model/Sana_Sprint_1.6B_1024px_teacher

           

🐱 Sana Model Card

This model serves as the Teacher in the figure below. It's not a few-step generative model but a fine-tuned diffusion model with (1) Dense Timestep Embedding and (2) QK Normalization discussed in the SANA-Sprint paper. Few-step generative models can be found in HF repo. Source code is available at https://github.com/NVlabs/Sana.

Training Pipeline

Model Description

• Developed by: NVIDIA, Sana
• Model type: Teacher model for One-Step Diffusion with Continuous-Time Consistency Distillation
• Model size: 1.6B parameters
• Model precision: torch.bfloat16 (BF16)
• Model resolution: This model is developed to generate 1024px based images with multi-scale heigh and width.
• License: 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.
• Model Description: This is 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 encoders (Gemma2-2B-IT) and one 32x spatial-compressed latent feature encoder (DC-AE).
• Resources for more information: Check out our GitHub Repository and the SANA-Sprint report on arXiv.

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 MIT Han-Lab provides free SANA-Sprint inference.

• Repository: https://github.com/NVlabs/Sana
• Demo: https://nv-sana.mit.edu/sprint

Uses

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.

Excluded uses are described below.

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.

Limitations and Bias

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.