Maintainer: hollowstrawberry

Total Score


Last updated 5/27/2024


Model LinkView on HuggingFace
API SpecView on HuggingFace
Github LinkNo Github link provided
Paper LinkNo paper link provided

Create account to get full access


If you already have an account, we'll log you in

Model overview

The holotard model is a set of AI models created by the maintainer hollowstrawberry. It is a collection of models that have been fine-tuned on various datasets related to anime and vtuber characters. The model is intended to be used with the stable-diffusion-webui tool, which provides an interface for generating images using AI models.

The model includes several specific checkpoints, including HeavenOrangeVtubers_hll4_final, AOM3_hll4_final, AOM2hard_hll4_final, and Grapefruit4.1_hll4_final. These models have been fine-tuned on various vtuber and anime-related datasets, and can be used to generate images with a distinct anime-inspired style.

Model inputs and outputs

The holotard model is an image-to-image AI model, meaning it takes an input image and generates a new image based on that input. The model can be used to create new anime-style images, or to modify existing images to have a more anime-inspired look and feel.


  • Input image: The model takes an image as input, which can be either a real photograph or a previously generated image.
  • Prompts: The model accepts textual prompts that describe the desired output image, such as specific characters, settings, or visual styles.
  • Loras: The model can also make use of Loras, which are additional machine learning models that can be used to apply specific visual styles or attributes to the output image.


  • Output image: The model generates a new image based on the input image and the provided prompts and Loras. The output image will have a distinct anime-inspired style, with characters, settings, and visual elements that match the input prompts.


The holotard model is capable of generating high-quality anime-style images with a wide range of characters, settings, and visual styles. The model has been fine-tuned on a variety of anime and vtuber-related datasets, and can generate images that capture the distinctive look and feel of these genres.

Some key capabilities of the holotard model include:

  • Generating images of anime-style characters, both individual and in group settings
  • Creating images with a range of different anime-inspired visual styles, including various art and animation techniques
  • Combining multiple elements, such as characters, settings, and objects, into cohesive and visually striking compositions
  • Applying Loras to modify and enhance the visual style of the output images

What can I use it for?

The holotard model can be used for a variety of creative and artistic projects, such as:

  • Generating concept art or illustrations for anime-inspired stories, games, or other media
  • Creating custom anime-style avatars or characters for use in various online platforms or applications
  • Enhancing and modifying existing images to have a more anime-inspired look and feel
  • Experimenting with different visual styles and techniques within the anime genre

Additionally, the model could potentially be used for commercial or professional applications, such as:

  • Developing anime-inspired assets or visuals for use in video games, films, or other media productions
  • Creating custom anime-style content or artwork for marketing, advertising, or branding purposes
  • Providing a tool for artists and designers to explore and experiment with anime-inspired styles and techniques

Things to try

When using the holotard model, there are several things you can experiment with to get the most out of the model:

  • Exploring different Loras: The model supports the use of Loras, which can be used to apply specific visual styles or attributes to the output images. Try using different Loras to see how they affect the final result.
  • Combining prompts and Loras: The model can be used in conjunction with textual prompts that describe the desired output. Try combining these prompts with the use of Loras to see how the model can create unique and compelling anime-inspired images.
  • Adjusting model parameters: The stable-diffusion-webui tool provides a range of parameters that can be adjusted to fine-tune the model's behavior, such as the number of inference steps, the sampling method, and the seed value. Experiment with these parameters to see how they affect the quality and style of the output images.
  • Iterating on the output: The model can be used to generate multiple iterations of an image, with each iteration building on the previous one. Try using the model to refine and improve the output over multiple generations.

By experimenting with the holotard model and the various tools and techniques available, you can unlock the full potential of this powerful AI-powered image generation system.

This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!

Related Models




Total Score


The Personal_Lora_collections model is a set of Lora (Low-Rank Adaptation) models created by the maintainer ikuseiso. Lora models are specialized AI models that can be used in conjunction with a base Stable Diffusion model to fine-tune its outputs for specific styles or characters. This collection includes Loras for a variety of anime-inspired characters and styles, such as vampy V3, vergil_devil_may_cry, sky_striker_ace_-_raye, and suletta_mercury. Similar models include loliDiffusion, which focuses on improving the generation of loli characters, and the sd-nai-lora-index, which is an index of various NovelAI-related Lora works. Model inputs and outputs Inputs Textual prompts describing the desired character or style Stable Diffusion base model Outputs Images of the specified character or style, generated using the Stable Diffusion model fine-tuned with the selected Lora Capabilities The Personal_Lora_collections model can generate a variety of anime-inspired characters and styles, ranging from vampy and gothic aesthetics to more heroic or magical girl-like designs. The maintainer notes that the model may be slightly overfitted, so adjusting the weights and adding additional prompts for things like hair or eye color can help improve the results. What can I use it for? The Personal_Lora_collections model can be used to create illustrations, concept art, or other visual assets featuring anime-inspired characters and styles. These could be used for personal projects, fan art, or even commercial applications like game or comic book development. The maintainer provides instructions for using the Lora models with the Stable Diffusion Web UI, making it accessible to a wide range of creators. Things to try One interesting aspect of the Personal_Lora_collections model is the maintainer's recommendation to adjust the weights of the Lora models, typically in the range of 0.6-0.8, to balance the fine-tuning and prevent overfitting. Experimenting with different weight values and prompts can help users find the right balance for their desired outputs. Additionally, trying out the various character-specific Loras, such as vergil_devil_may_cry or suletta_mercury, can showcase the model's versatility in capturing different anime-inspired styles and designs.

Read more

Updated Invalid Date



Total Score


The LoraByTanger model is a collection of Lora models created by Tanger, a Hugging Face community member. The main focus of this model library is on Genshin Impact characters, but it is planned to expand to more game and anime characters in the future. Each Lora folder contains a trained Lora model, a test image generated using the "AbyssOrangeMix2_hard.safetensors" model, and a set of additional generated images. Model inputs and outputs Inputs Text prompts describing the desired character or scene, which the model uses to generate images. Outputs High-quality, detailed anime-style images based on the input text prompt. Capabilities The LoraByTanger model is capable of generating a wide variety of anime-inspired images, particularly focused on Genshin Impact characters. The model can depict characters in different outfits, poses, and settings, showcasing its versatility in generating diverse and aesthetically pleasing outputs. What can I use it for? The LoraByTanger model can be useful for a variety of applications, such as: Creating custom artwork for Genshin Impact or other anime-inspired games and media. Generating character designs and illustrations for personal or commercial projects. Experimenting with different styles and compositions within the anime genre. Providing inspiration and reference material for artists and illustrators. Things to try One key aspect to explore with the LoraByTanger model is the impact of prompt engineering and the use of different tags or modifiers. By adjusting the prompt, you can fine-tune the generated images to match a specific style or character attributes. Additionally, experimenting with different Lora models within the collection can lead to unique and varied outputs, allowing you to discover the nuances and strengths of each Lora.

Read more

Updated Invalid Date



Total Score


The loliDiffusion model is a text-to-image diffusion model created by JosefJilek that aims to improve the generation of loli characters compared to other models. This model has been fine-tuned on a dataset of high-quality loli images to enhance its ability to generate this specific style. Similar models like EimisAnimeDiffusion_1.0v, Dreamlike Anime 1.0, waifu-diffusion, and mo-di-diffusion also focus on generating high-quality anime-style images, but with a broader scope beyond just loli characters. Model Inputs and Outputs Inputs Textual Prompts**: The model takes in text prompts that describe the desired image, such as "1girl, solo, loli, masterpiece". Negative Prompts**: The model also accepts negative prompts that describe unwanted elements, such as "EasyNegative, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, multiple panels, aged up, old". Outputs Generated Images**: The primary output of the model is high-quality, anime-style images that match the provided textual prompts. The model is capable of generating images at various resolutions, with recommendations to use standard resolutions like 512x768. Capabilities The loliDiffusion model is particularly skilled at generating detailed, high-quality images of loli characters. The prompts provided in the model description demonstrate its ability to create images with specific features like "1girl, solo, loli, masterpiece", as well as its flexibility in handling negative prompts to improve the generated results. What Can I Use It For? The loliDiffusion model can be used for a variety of entertainment and creative purposes, such as: Generating personalized artwork and illustrations featuring loli characters Enhancing existing anime-style images with loli elements Exploring and experimenting with different loli character designs and styles Users should be mindful of the sensitive nature of loli content and ensure that any use of the model aligns with applicable laws and regulations. Things to Try Some interesting things to try with the loliDiffusion model include: Experimenting with different combinations of positive and negative prompts to refine the generated images Combining the model with other text-to-image or image-to-image models to create more complex or layered compositions Exploring the model's performance at higher resolutions, as recommended in the documentation Comparing the results of loliDiffusion to other anime-focused models to see the unique strengths of this particular model Remember to always use the model responsibly and in accordance with the provided license and guidelines.

Read more

Updated Invalid Date




Total Score


The lora-training model is a collection of various LoRA (Low-Rank Adaptation) models trained by maintainer khanon on characters from the mobile game Blue Archive. LoRA is a technique used to fine-tune large language models like Stable Diffusion in an efficient and effective way. This model library includes LoRAs for characters like Arona, Chise, Fubuki, and more. The preview images demonstrate the inherent style of each LoRA, generated using ControlNet with an OpenPose input. Model inputs and outputs Inputs Images of characters from the mobile game Blue Archive Outputs Stylized, high-quality images of the characters based on the specific LoRA model used Capabilities The lora-training model allows users to generate stylized, character-focused images based on the LoRA models provided. Each LoRA has its own unique artistic style, allowing for a range of outputs. The maintainer has provided sample images to showcase the capabilities of each model. What can I use it for? The lora-training model can be used to create custom, stylized images of Blue Archive characters for a variety of purposes, such as fan art, character illustrations, or even asset creation for games or other digital projects. The LoRA models can be easily integrated into tools like Stable Diffusion to generate new images or modify existing ones. Things to try Experiment with different LoRA models to see how they affect the output. Try combining multiple LoRAs or using them in conjunction with other image generation techniques like ControlNet. Explore how the prompts and settings affect the final image, and see if you can push the boundaries of what's possible with these character-focused LoRAs.

Read more

Updated Invalid Date