pipeline_tag: text-to-image
license: apache-2.0
base_model:
- neta-art/Neta-Lumina
- Alpha-VLLM/Lumina-Image-2.0
tags:
- stable-diffusion
- text-to-image
- comfyui
- diffusion-single-file
NetaYume Lumina Image v2.0
This model is based on Lumina-Image-2.0, which is a DIT model with 2 billions parameter flow-based diffusion transformer. For more information, visit here.
I. Introduction
NetaYume Lumina is a text-to-image model fine-tuned from Neta Lumina, a high-quality anime-style image generation model developed by Neta.art Lab. It builds upon Lumina-Image-2.0, an open-source base model released by the Alpha-VLLM team at Shanghai AI Laboratory.
This model was trained with the goal of not only generating realistic human images but also producing high-quality anime-style images. Despite being fine-tuned on a specific dataset, it retains a significant amount of knowledge from the base model.
Key Features:
High-Quality Anime Generation: Generates detailed anime-style images with sharp outlines, vibrant colors, and smooth shading.
Improved Character Understanding: Better captures characters, especially those from the Danbooru dataset, resulting in more coherent and accurate character representations.
Enhanced Fine Details: Accurately generates accessories, clothing textures, hairstyles, and background elements with greater clarity.
The file NetaYume_Lumina_v2_all_in_one.safetensors is an all-in-one file that contains the necessary weights for the VAE, text encoder, and image backbone to be used with ComfyUI.
- Model Components & Training Details
- Text Encoder: Pre-trained Gemma-2-2b
- Variational Autoencoder: Pre-trained Flux.1 dev's VAE
- Image Backbone: Fine-tune NetaLumina's Image Backbone
- Suggestion
System Prompt: This help you generate your desired images more easily by understanding and aligning with your prompts.
For anime-style images using Danbooru tags:
You are an assistant designed to generate anime images based on textual prompts.
You are an assistant designed to generate high-quality images based on user prompts and danbooru tags.
Recommended Settings
- CFG: 4–8
- Sampling Steps: 40-50
- Sampler:
- Euler a (with scheduler: normal)
- res_multistep (with scheduler: linear_quadratic)
- Acknowledgments
- narugo1992 – for the invaluable Danbooru dataset
- Alpha-VLLM - for creating the a wonderful model!
- Neta Lumina and his team – for openly sharing a wonderful model.