huggingface/diffusers

Health v2 · maintenance only

Adoption not part of the score

34385 stars · 7268 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Hugging Face Diffusers is a Python library providing state-of-the-art pretrained diffusion models for generating images, videos, and audio in PyTorch. It offers ready-to-use DiffusionPipeline APIs for inference, interchangeable noise schedulers, and modular building blocks for training custom diffusion systems.

Use cases

• generate images from text prompts with stable diffusion
• run text-to-video generation in python
• train or finetune a diffusion model with LoRA
• convert text to audio or speech via audio diffusion models
• build a custom image generation pipeline with custom schedulers
• run large diffusion models on limited GPU memory with cpu offloading
• do image-to-image editing with diffusion models

When to choose

• you want a well-maintained, widely-adopted library for diffusion model inference or training in PyTorch
• you need to mix and match model components, schedulers, and adapters like LoRA
• you need memory optimizations like CPU offloading and quantization for large models
• you want access to the latest models (Stable Diffusion, Flux, Qwen-Image) from the Hugging Face Hub

When to avoid

• you need a one-click GUI or hosted API rather than a Python library
• your project is not Python/PyTorch based
• you only need simple image classification or non-generative vision tasks

Facets

library · maturity stable

machine-learning deep-learning image-processing audio-processing video-processing gpu-computing deep-learning machine-learning artificial-intelligence image-processing python cross-platform diffusion-models stable-diffusion text-to-image text-to-video pytorch generative-ai lora hugging-face video audio gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="huggingface/diffusers")