This is the SVDQ quantization version of the Project0 model, a text-to-image model created based on Flux Dev and Flux Krea. The model offers various quantization formats, optimized for different GPU architectures, including INT4 and FP4 quantization versions, suitable for both Blackwell and non-Blackwell GPU users.

MultimodalDiffusersEnglish

PixelWave is a text-to-image model developed based on Flux.1. It has been optimized through Nunchaku quantization (SVDQ), providing an efficient image generation solution for users with different GPU configurations.

Computer VisionDiffusersEnglish

Outfit-Cut is an adapter for FLUX.1-Kontext-dev of black-forest-lab, specifically designed to accurately extract specified clothing items from images. This model has been trained on 200 pairs of images and can place the clothing on a clean, solid-color background, presenting it in a professional product photography style while preserving the original texture, color, shape, and details of the clothing.

Computer VisionDiffusers

SRPO is a human preference alignment method for diffusion models. Through Direct-Align technology and semantic relative preference optimization, it significantly improves the realism and aesthetic quality of the FLUX.1-dev model, and solves the problems of high computational cost in multi-step denoising and dependence on offline reward fine-tuning.

Computer VisionDiffusers

Chroma1-HD is a text-to-image generation base model based on the FLUX.1-schnell architecture, with 8.9 billion parameters, and uses the Apache 2.0 open-source license. This model is specifically designed for fine-tuning, providing a powerful image generation foundation for developers and researchers, and supporting the creation of various specialized visual content.

Computer VisionDiffusers

FLUX.1 [dev] is a rectified flow transformer with 12 billion parameters, specifically designed for text-to-image generation. It excels in image generation quality, second only to FLUX.1 [pro], with excellent prompt following ability and an efficient training method, providing strong support for scientific research and creation.

MultimodalDiffusersEnglish

FLUX.1 Krea [dev] is a rectified flow transformer with 12 billion parameters, capable of generating high-quality images based on text descriptions. This model focuses on aesthetic photography and excels in image quality and instruction following. It uses guided distillation for efficient training and opens its weights to promote scientific research and artistic creation.

MultimodalDiffusersEnglish

FLUX.1 Krea [dev] is an advanced rectified flow transformer model with 12 billion parameters, specifically designed to generate high-quality images based on text descriptions. This model excels in the field of aesthetic photography, has excellent prompt-following ability, is trained using guided distillation technology, and its weights are open for non-commercial use.

Computer VisionDiffusersEnglish

Chroma1-Base is a text-to-image basic model with 8.9 billion parameters, built on the FLUX.1-schnell architecture and licensed under the Apache 2.0 license. This model is designed for fine-tuning, providing a powerful and neutral foundation for developers, researchers, and artists to create specialized generative models.

MultimodalDiffusers

The Nunchaku quantized version of the Colossus Project Flux, designed to generate high-quality images based on text prompts. This model minimizes performance loss while optimizing inference efficiency.

Computer VisionDiffusersEnglish

The Nunchaku-quantized FLUX.1-dev model is used to generate high-quality images based on text prompts and optimizes inference efficiency.

Computer VisionDiffusersEnglish

This is a LyCORIS adapter based on FLUX.1-schnell, specifically designed for text-to-image generation tasks. This model is fine-tuned using LoRA technology. During the training process, no validation prompts were used, and the text encoder was not trained. The text encoder of the base model can be reused for inference.

Computer VisionDiffusers

FLUX.1 Krea [dev] is a rectified flow transformer with 12 billion parameters, specifically designed to generate high-quality images based on text descriptions. It excels in output quality, instruction following, etc., focuses on aesthetic photography, adopts an efficient guided distillation training method, and opens its weights to promote scientific research and creative work.

Computer VisionDiffusersEnglish

Provide LoRA adapters with more than 20 artistic and cartoon styles for the FLUX.1 Kontext model to achieve high-quality image-to-image generation.

Computer VisionDiffusersEnglish

FLUX.1 [dev] is a rectified flow transformer with 12 billion parameters, capable of generating high-quality images based on text descriptions, providing powerful image generation capabilities for developers and creative professionals.

MultimodalGgufEnglish

An experimental diffusion model based on FLUX.1-Kontext-dev, focusing on image-to-image conversion tasks. Quantization processing has been completed, significantly improving inference performance. The video memory usage is about 10GB, and only 6 - 8 steps are required for inference.

Computer VisionDiffusers

The FLUX.1-Kontext-dev version processed by Nunchaku quantization can edit images according to text instructions, optimizing inference efficiency while minimizing performance loss.

Computer VisionDiffusersEnglish

The Nunchaku quantized version of FLUX.1-Fill-dev, capable of filling regions in existing images based on text descriptions, achieving efficient inference through quantization processing.

Computer VisionDiffusersEnglish

The Nunchaku-quantized FLUX.1-schnell model is used to generate high-quality images based on text prompts and has been optimized for efficient inference.

Computer VisionDiffusersEnglish

The quantized version of FLUX.1-Depth-dev provided by the Nunchaku team can generate images based on text descriptions while preserving the structure of the input image, optimizing the inference efficiency.

Computer VisionDiffusersEnglish