Coreml-community
Models by this creator
⛏️
coreml-ChilloutMix
93
The coreml-ChilloutMix model is a Core ML-converted version of the Chilloutmix model, which was originally trained on a dataset of "wonderful realistic models" and merged with the Basilmix model. This model is designed for generating realistic images of Asian girls in NSFW poses. The maintainer, the coreml-community, has provided several versions of the model, including split_einsum and original versions, as well as custom resolution and VAE-embedded variants. The model was converted to Core ML for use on Apple Silicon devices, with instructions available for converting other Stable Diffusion models to the Core ML format. Similar models include chilloutmix, chilloutmix-ni, and ambientmix from other creators. Model inputs and outputs Inputs Text prompts to describe the desired image Outputs Realistic, high-quality images of Asian girls in NSFW poses Capabilities The coreml-ChilloutMix model is capable of generating detailed, realistic images of Asian girls in a variety of NSFW poses and scenarios. The model has been trained on a dataset of "wonderful realistic models" and can produce images with a high level of detail and naturalism. What can I use it for? The coreml-ChilloutMix model could be useful for NSFW content creators or artists looking to generate realistic images of Asian girls. The model's capabilities could be leveraged for a variety of projects, such as character design, illustrations, or adult-themed artwork. However, users should be aware of the model's NSFW nature and ensure that any use of the model aligns with relevant laws and ethical considerations. Things to try One interesting aspect of the coreml-ChilloutMix model is its ability to generate realistic Asian features and skin textures. Users could experiment with prompts that focus on these elements, such as "highly detailed skin texture" or "beautifully rendered Asian facial features." Additionally, the model's compatibility with various compute unit options, including the Neural Engine, could be explored to optimize performance on different hardware.
Updated 5/28/2024
🤯
coreml-stable-diffusion-2-1-base
74
The coreml-stable-diffusion-2-1-base model is a Core ML converted version of the Stable Diffusion v2-1-base model developed by Stability AI. It is a latent diffusion model that can be used to generate and modify images based on text prompts. The model was fine-tuned from the stable-diffusion-2-base model with an additional 220k steps, and has improved performance compared to the base model. Model inputs and outputs The coreml-stable-diffusion-2-1-base model takes text prompts as input and generates corresponding images as output. The text prompts are encoded using a fixed, pretrained text encoder (OpenCLIP-ViT/H), and the generated images are produced in the latent space of the model. Inputs Text prompts**: Short text descriptions that describe the desired image to generate. Outputs Generated images**: The model outputs images that correspond to the provided text prompts. Capabilities The coreml-stable-diffusion-2-1-base model can be used to generate a wide variety of images based on text prompts, including scenes, objects, and abstract concepts. The model has been fine-tuned to improve its performance compared to the base Stable Diffusion v2 model, and can produce higher-quality and more detailed images. What can I use it for? The coreml-stable-diffusion-2-1-base model is intended for research purposes, such as understanding the limitations and biases of generative models, generating artworks, and developing creative tools. It could also be used in educational settings or for personal creative projects. However, the model should not be used to intentionally create or disseminate images that are harmful, offensive, or propagate stereotypes. Things to try One interesting thing to try with the coreml-stable-diffusion-2-1-base model is to experiment with different text prompts and see how the generated images vary. You could also try using the model's capabilities to assist with creative tasks, such as designing album covers or exploring new artistic styles. Additionally, you could investigate the model's limitations, such as its inability to render legible text or accurately depict faces and people.
Updated 5/28/2024
⛏️
ControlNet-Models-For-Core-ML
56
The ControlNet-Models-For-Core-ML is a collection of ControlNet models converted to the Apple CoreML format by the coreml-community maintainer. ControlNet is a neural network structure that allows controlling pretrained large diffusion models like Stable Diffusion by adding extra conditioning inputs. These CoreML models are specifically designed for use with Swift apps like MOCHI DIFFUSION or the SwiftCLI, and are not compatible with Python-based Diffusers pipelines. The models in this repository include both "Original" and "Split-Einsum" versions, all built for Stable Diffusion v1.5. They feature various conditioning inputs such as Canny edge detection, Midas depth estimation, HED edge detection, MLSD line detection, surface normal estimation, OpenPose pose detection, scribbles, and semantic segmentation. These conditioning inputs can be used to guide and control the image generation process. Model inputs and outputs Inputs Conditioning Image**: An image that provides additional input information to guide the image generation process, such as edge maps, depth maps, poses, etc. Text Prompt**: A text description that specifies the desired output image. Outputs Generated Image**: The final output image generated by the model, based on the provided text prompt and conditioning image. Capabilities The ControlNet-Models-For-Core-ML models excel at generating images that adhere to specific visual constraints or guidelines, such as incorporating detailed edge information, depth cues, or semantic segmentation. This allows for more precise control over the generated imagery, enabling users to create images that closely match their desired visual characteristics. What can I use it for? These ControlNet models are well-suited for various creative and artistic applications, such as generating concept art, illustrations, or visualizations that require a high degree of control over the output. Developers of Swift apps focused on image generation or manipulation can leverage these models to offer users more advanced capabilities beyond standard text-to-image generation. Things to try Experiment with different conditioning inputs and prompts to see how the models respond. Try using edge maps, depth information, or pose data to guide the generation of specific types of images, such as architectural renderings, character designs, or product visualizations. Additionally, explore the differences between the "Original" and "Split-Einsum" versions to see how they impact the quality and performance of the generated outputs.
Updated 5/28/2024
🏷️
coreml-anything-v3.0
48
The coreml-anything-v3.0 model is a text-to-image generation model developed by the coreml-community that has been converted to Core ML for use on Apple Silicon devices. This model is based on the Anything V3.0 model, which is a latent diffusion model intended to produce high-quality, highly detailed anime-style images with just a few prompts. Like other anime-style Stable Diffusion models, it also supports Danbooru tags to generate images. Similar models include the coreml-stable-diffusion-2-1-base model, which is a Core ML-converted version of the Stable Diffusion v2.1-base model, and the anything-v3-1 model, which is a third-party continuation of the Anything V3.0 model with a fixed VAE and CLIP model. Model inputs and outputs Inputs Text prompts**: The model takes text prompts as input, which can include Danbooru tags and other descriptive text to guide the image generation process. Outputs Images**: The model outputs high-quality, highly detailed anime-style images based on the provided text prompts. Capabilities The coreml-anything-v3.0 model is capable of generating a wide variety of anime-style images, from character portraits to detailed scenes. It can produce images with a high level of visual complexity and realism, making it a powerful tool for artists, designers, and hobbyists interested in creating anime-inspired content. What can I use it for? The coreml-anything-v3.0 model can be used for a variety of applications, including: Creative projects**: Use the model to generate unique, anime-style images for use in illustrations, concept art, or other creative endeavors. Prototyping and ideation**: Quickly generate images to explore ideas and concepts, or to create visual references for further development. Educational and research purposes**: Investigate the model's capabilities, limitations, and potential biases as part of research on generative AI models. Artistic tools**: Integrate the model into applications or workflows to assist artists in their creative process. Things to try One interesting thing to try with the coreml-anything-v3.0 model is experimenting with different prompt styles and combinations of Danbooru tags. The model is designed to work well with these types of descriptive prompts, so trying out various tag combinations can lead to unique and unexpected results. Additionally, leveraging the model's strengths in generating detailed anime-style imagery can be a great way to explore new creative avenues and push the boundaries of what's possible with text-to-image generation.
Updated 9/6/2024