v1.0.0-beta.22

• Added SD Turbo model(Pytorch/OpenVINO) support (1-step inference)
• Added SD Turbo/SDXL Turbo models image to image support

v1.0.0-beta.21

• Added SDXL Turbo OpenVINO support (2.5 seconds to generate image)

v1.0.0-beta.20

• Added support for 1-step ultra-fast text-to-image generation (SDXL Turbo)

v1.0.0-beta.19

• Fixed tiny decoder issue in the image-to-image pipeline

v1.0.0-beta.18

• Added image-to-image support (Use WebUI)
• Improved WebUI
• OpenVINO image-to-image support

v1.0.0-beta.17

• Configurable lcm models, check configs/lcm-models.txt file
• Fixed Qt GUI layout spacing problem

v1.0.0-beta.16

• Added 2 steps inference for LCM-LoRA workflow

v1.0.0-beta.15

• Fast 2,3 steps inference
• Lcm-Lora fused models for faster inference
• Added real-time text to image generation on CPU (Experimental)
• Fixed DPI scale issue
• Fixed SDXL tiny auto decoder issue
• Supports integrated GPU(iGPU) using OpenVINO (export DEVICE=GPU)
• 5.7x speed using OpenVINO(steps: 2,tiny autoencoder)

v1.0.0-beta.13

• Added support for custom models for OpenVINO (LCM-LoRA baked)
• Added negative prompt support for OpenVINO models (Set guidance >1.0)
• 2x faster inference on CPU

v1.0.0-beta.12

• Added SDXL, SSD - 1B LCM models
• Added LCM-LoRA support, which works well for fine-tuned Stable Diffusion model 1.5 or SDXL models
• Added negative prompt support in LCM-LoRA mode
• LCM-LoRA models can be configured using a text configuration file