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

A training model based on flux learning, capable of accurately capturing the delicate techniques and stylistic elements of artists like Renoir.

Computer VisionDiffusers

DanTagGen is a tag generation model trained from scratch based on the LLaMA architecture with 400 million parameters, specifically designed to generate relevant tags for art descriptions. This model is trained on the Danbooru dataset and can automatically generate suitable tags based on the input information such as artists, characters, and copyrights.

Natural Language ProcessingTransformersMultiple Languages

A latent diffusion model trained on works by Japanese artists, supporting non-square resolution generation and Danbooru tag input

Computer VisionDiffusers