animagine-xl-4.0 by Cagliostro Labs: Open-Weight Model

Runs On

What it takes to serve animagine-xl-4.0 (2.6B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

Precision	Weights	Memory needed	Cheapest setup	Per hour	Also fits
16-bit	5.1 GB	6.2 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
8-bit	2.6 GB	3.1 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
4-bit	1.3 GB	1.5 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

SAVRN's Notes on animagine-xl-4.0

The download is bigger than the model. Cagliostro Labs ships 22 files totaling 20.8 GB, yet the 2.6B parameters you actually serve take 5.1 GB at 16-bit and need 6.2 GB of memory in use. This anime-styled text-to-image model was retrained from Stable Diffusion XL 1.0 on 8.4M images, and it fits on one 192 GB MI300X at $1.85 per hour on-demand with most of the card left idle. At 8-bit memory drops to 3.1 GB, so one card can hold many parallel workers.

Open RAIL++ is where a buyer should slow down. Commercial use is permitted but conditional on use-based restrictions, and those restrictions must be passed to your own downstream users, so the terms travel with every product built on it. Also note the lineage: the file lists it as derived from stabilityai/stable-diffusion-xl-base-1.0, the knowledge cutoff is January 7, 2025, and the last update was February 13, 2025.

SAVRN Research, 2026-09-18

Model Card

By Cagliostro Labs, published under openrail++, revision 2b7c1b397761.

Overview

Animagine XL 4.0, also stylized as Anim4gine, is the ultimate anime-themed finetuned SDXL model and the latest installment of Animagine XL series. Despite being a continuation, the model was retrained from Stable Diffusion XL 1.0 with a massive dataset of 8.4M diverse anime-style images from various sources with the knowledge cut-off of January 7th 2025 and finetuned for approximately 2650 GPU hours. Similar to the previous version, this model was trained using tag ordering method for the identity and style training. With the release of Animagine XL 4.0 Opt (Optimized), the model has been further refined with an additional dataset, improving stability, anatomy accuracy, noise reduction, color saturation, and overall color accuracy. These enhancements make Animagine XL 4.0 Opt more consistent and visually appealing while maintaining the signature quality of the series.

Changelog

Read the full model card (1,511 words)

Identity and Version

Repository: cagliostrolab/animagine-xl-4.0
Publisher: Cagliostro Labs
Task: Text to image
Modality: Image
Library: diffusers
Parameters: 2.6B parameters
Languages: en
Revision: 2b7c1b397761bf5bd3cc42e5b39ec99314a75a96
First published: 2025-01-10
Last updated: 2025-02-13

Files and Weights

22 files, 20.8 GB in total. The weights are 6 files totalling 20.8 GB in safetensors.

Weights6 files · 20.8 GB

Configuration8 files · 5.9 KB

Tokenizer6 files · 3.2 MB

Documentation1 file · 15.5 KB

Repository1 file · 1.5 KB

Every file

File	Type	Size	SHA-256
animagine-xl-4.0-opt.safetensors	Weights	6.9 GB	6327eca98bfb
animagine-xl-4.0.safetensors	Weights	6.9 GB	1d5b43ff75b6
text_encoder/model.safetensors	Weights	246.1 MB	73528dedc471
text_encoder_2/model.safetensors	Weights	1.4 GB	1669a15b906d
unet/diffusion_pytorch_model.safetensors	Weights	5.1 GB	3e16857a3a19
vae/diffusion_pytorch_model.safetensors	Weights	167.3 MB	6353737672c9
model_index.json	Configuration	671 B	—
scheduler/scheduler_config.json	Configuration	613 B	—
text_encoder/config.json	Configuration	560 B	—
text_encoder_2/config.json	Configuration	570 B	—
tokenizer/special_tokens_map.json	Configuration	472 B	—
tokenizer_2/special_tokens_map.json	Configuration	460 B	—
unet/config.json	Configuration	1.8 KB	—
vae/config.json	Configuration	782 B	—
README.md	Documentation	15.5 KB	—
.gitattributes	Repository	1.5 KB	—
tokenizer/merges.txt	Tokenizer	524.6 KB	—
tokenizer/tokenizer_config.json	Tokenizer	704 B	—
tokenizer/vocab.json	Tokenizer	1.1 MB	—
tokenizer_2/merges.txt	Tokenizer	524.6 KB	—
tokenizer_2/tokenizer_config.json	Tokenizer	855 B	—
tokenizer_2/vocab.json	Tokenizer	1.1 MB	—

License and Download

License: openrail++
Access: Open weights, no gate
Download size: 20.8 GB

Download from Cagliostro Labs

Released by Cagliostro Labs through its official repository on Hugging Face.

Built From

Derived from stabilityai/stable-diffusion-xl-base-1.0

Memory Requirements

Precision	Weights in memory
As published	20.8 GB
16-bit	5.1 GB
8-bit	2.6 GB
4-bit	1.3 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Compare animagine-xl-4.0

animagine-xl-4.0 vs stable-diffusion-xl-base-1.0

Questions About animagine-xl-4.0

How much GPU memory does animagine-xl-4.0 need?

About 6.2 GB at 16-bit and 1.5 GB at 4-bit: the weights (2.6B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run animagine-xl-4.0 on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use animagine-xl-4.0 commercially?

Yes, with conditions. animagine-xl-4.0 is released under Open RAIL++ License. Open RAIL++ permits use, including commercial use, subject to the use-based restrictions listed in the license, which must be passed on to downstream users.

Similar Models

Model · Text to image

stable-diffusion-xl-base-1.0

Stability AI

SDXL consists of an ensemble of experts pipeline for latent diffusion: In a first step, the base model is used to generate (noisy) latents, which are then further processed with a refinement model (available here: https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0/) specialized for the final denoising steps. Note that the base model can be used as a standalone module. Alternatively, we can use a two-stage pipeline as follows: First, the base model is used to generate latents of the desired output size. In the second step, we use a specialized high-resolution model and apply a technique called SDEdit (https://arxiv.org/abs/2108.01073, also known as "img2img") to the latents…

Open weights openrail++ 2.6B parameters diffusers

View model

Model · Text to image

sdxl-turbo

Stability AI

SDXL-Turbo is a fast generative text-to-image model that can synthesize photorealistic images from a text prompt in a single network evaluation. A real-time demo is available here: http://clipdrop.co/stable-diffusion-turbo Please note: For commercial use, please refer to https://stability.ai/license. SDXL-Turbo is a distilled version of SDXL 1.0, trained for real-time synthesis. SDXL-Turbo is based on a novel training method called Adversarial Diffusion Distillation (ADD) (see the technical report), which allows sampling large-scale foundational image diffusion models in 1 to 4 steps at high image quality. This approach uses score distillation to leverage large-scale off-the-shelf image…

Open weights other 2.6B parameters diffusers

View model

Model · Text to image

playground-v2.5-1024px-aesthetic

Playground

This repository contains a model that generates highly aesthetic images of resolution 1024x1024, as well as portrait and landscape aspect ratios. You can use the model with Hugging Face Diffusers. Playground v2.5 is a diffusion-based text-to-image generative model, and a successor to Playground v2. Playground v2.5 is the state-of-the-art open-source model in aesthetic quality. Our user studies demonstrate that our model outperforms SDXL, Playground v2, PixArt-α, DALL-E 3, and Midjourney 5.2. For details on the development and training of our model, please refer to our blog post and technical report. Install diffusers >= 0.27.0 and the relevant dependencies. - The pipeline uses the…

Open weights other 2.6B parameters diffusers

View model

Model · Text to image

animagine-xl-3.1

Cagliostro Labs

/ FIXED: Changed from 50% to 33.33% because there are 3 columns / margin-bottom: 1em; / Added small margin for spacing between stacked images / font-weight: bold; / Corrected 'font-style: bold' to 'font-weight: bold' / } / FIXED: Added missing closing brace here /.overlay, Animagine XL 3.1 is an update in the Animagine XL V3 series, enhancing the previous version, Animagine XL 3.0. This open-source, anime-themed text-to-image model has been improved for generating anime-style images with higher quality. It includes a broader range of characters from well-known anime series, an optimized dataset, and new aesthetic tags for better image creation. Built on Stable Diffusion XL, Animagine XL 3.1…

Open weights openrail++ 2.6B parameters diffusers

View model

Model · Text to image

nova-furry-xl-il-v120-sdxl

John Smith

Original model is here. This model created by Crody.

Open weights other 2.6B parameters diffusers

View model

Model · Text to image

noobai-XL-1.1

Laxhar Dream Lab

This is an image generation model based on training from Illustrious-xl. It utilizes the latest full Danbooru and e621 datasets for training, with native tags caption. For quality tags, we evaluated image popularity through the following process: - Data normalization based on various sources and ratings. - Application of time-based decay coefficients according to date recency. - Ranking of images within the entire dataset based on this processing. Our ultimate goal is to ensure that quality tags effectively track user preferences in recent years. - Latest Danbooru images up to the training date(for v1.0,it mean approximately before 2024-10-23) - E621 images e621-2024-webp-4Mpixel dataset on…

Open weights other 2.6B parameters diffusers

View model