invoke-ai/InvokeAI
PROFESSIONAL AI ENGINE FOR VISUAL MEDIA
InvokeAI is a local web interface for generating and editing images and videos using diffusion models. Supports dozens of models, node-based workflows, canvas with inpainting, and model management.
// KEY FEATURES
- React web UI with local server
- Unified canvas with inpainting, outpainting, brushes
- Node-based workflows for custom generation pipelines
- Support for SD, Flux, Qwen, Ideogram and 20+ models
// DEEP DIVE: HOW IT WORKS
InvokeAI runs a local web server with a React interface for interacting with diffusion models. Generation happens on user hardware: checkpoints (ckpt, diffusers, GGUF) are loaded, parameters configured, and pipelines executed. The UI combines a drawing canvas, node-based workflow editor, and gallery with metadata.
It stands out with a node-based workflow approach: users build custom pipelines from nodes and share them. Supports a vast model range — from SD 1.5 to Flux.2, Qwen, Ideogram, ERNIE; some models (Nano Banana, GPT Image, Wan) are API-only. Includes model and embedding managers, upscalers, SAM/SAM2 segmentation.
Requires compatible hardware (GPU with VRAM for large models). Installed via launcher; runs on Linux, Windows, macOS (per repo topics). License allows commercial use. API models require provider keys.
// USE CASES
Best for: For artists and designers doing local image generation and editing
- Generate concept art and iteratively refine on canvas with inpainting
- Build node-based workflows for repeatable production generation pipelines
- Organize generation library with metadata for quick remixing
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram