Stable Diffusion 1.5

Generate high-quality images from textual descriptions by leveraging advanced deep learning techniques

Model Properties

Uses a Text Encoder to turn prompts into embeddings, an Image Decoder for latent representation, and the UNet Model in 20 steps. Supports batch processing (positive and negative prompts).

License name: CreativeML Open RAIL-M License
Number of parameters: 1B
Model size: 1.00 GB
Select device..

Technical details

Numerical Scheme: A8W8, channel-wise, symmetric, channel-wise, symmetric
Inference Api: CPP
Compiled Model:

Performance Metrics

Number Of Iterations 20
First Load Time In Sec 5.89
FPS 0.0625
Sub-Model Performance Metrics
Text Encoder
Output Tokens 77
Number Of Parameters 85.1M
Operations 13.9 GOPs per input token
FPS 16
Image Decoder
Output Resolution [512, 512, 3]
Number Of Parameters 49.5M
Operations 2500 GOPs per input token
FPS 0.45
UNet Model
Output Resolution [64, 64, 4]
Number Of Parameters 904.2M
Operations 840.5 GOPs per input token
FPS 1.9

Explore Related Models

Gen AI _ Stable Diffusion
GenAI Models
Qwen2-VL-2B-Instruct
Generate multimodal responses by interpreting both text and images, enabling vision-language understanding and content creation
Qwen2.5-Coder-1.5B-Instruct
GenAI Models
Qwen2.5-Coder-1.5B
Generate text responses to prompts, enabling natural language understanding, multilingual support, and code generation