Smalldiffusion

Visit Tool

smalldiffusion is an open-source coding & development tool that provides simple and readable code for training and sampling from diffusion models. It supports various models and parameterizations for ease of experimentation.

Claim this tool

1View

At a glance

Pricing

Open Source

Free tier

Yes

API

Skill level

Technical

About

What is smalldiffusion?

smalldiffusion is a lightweight, open-source Python library designed for training and sampling from diffusion and flow models. It prioritizes ease of experimentation, allowing developers and researchers to quickly train new models or develop novel samplers. The library supports a variety of models including MLP, U-Net, and DiT, and multiple parameterizations such as score-, flow-, or data-prediction. It offers dataset support for 2D toy datasets, pixel, and latent-space image datasets, with example training code for FashionMNIST, CIFAR10, and Imagenet. smalldiffusion also provides concise implementations of diffusion transformers and supports conditional training with classifier-free guidance, making it a versatile tool for those working with diffusion models.

Best used for

Ideal for developers and professors who need to quickly prototype and experiment with diffusion models, implement custom samplers, and train models on various datasets. Especially valuable for those seeking readable and extensible code for research and development in AI.

Common actions

train diffusion models

sample diffusion models

experiment with AI models

implement AI algorithms

face swapping"AI Agents"automated workflowopen-sourcecollaborationgithub copilotlow-code/no-codeworkflowsdeepfake

Capabilities

Key features

Lightweight diffusion library
Train/sample diffusion models
MLP, U-Net, DiT models
Multiple parameterizations
Dataset support
Conditional training
Accelerated sampling algorithms

Target Audience

developerprofessor

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What kind of diffusion models can I train with smalldiffusion?

smalldiffusion supports training and sampling from various diffusion and flow models, including MLP, U-Net, and Diffusion Transformer (DiT) architectures. It also allows for multiple parameterizations like score-, flow-, or data-prediction, offering flexibility for different research and development needs.

Does smalldiffusion support conditional generation?

Yes, smalldiffusion supports conditional training and sampling, including classifier-free guidance. This allows users to train models that can generate outputs based on specific conditions, such as class labels, as demonstrated in examples like the 2D tree dataset.

Can I use smalldiffusion with existing pretrained models like Stable Diffusion?

Yes, smalldiffusion's sampler is designed to work with any pretrained diffusion model. It provides a simple wrapper for Hugging Face Diffusers latent diffusion models, enabling sampling from pretrained models like Stable Diffusion with minimal code and allowing for easy experimentation with sampler parameters.

Trending

Subcategories trending in Coding & Development

Code Assistants DevOps & Infrastructure No-Code / Low-Code Testing & QA Backend & APIs Prompt Engineering

Trending

Also listed in

This tool also appears in

Research & Education › Academic Research AI Agents & Automation › AI Frameworks & Infra Research & Education › Scientific Computing

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce