InvokeAI
Free, open-source, self-hosted creative engine for AI image generation with a layered canvas and node-based workflows.
At a Glance
Free, self-hosted, Apache 2.0 licensed Invoke creative engine for AI image generation.
Engagement
Available On
Alternatives
Listed Nov 2026
About InvokeAI
Invoke is an open-source, Apache 2.0 licensed creative engine for AI image generation that runs locally on the user's own hardware. It provides a web-based UI with a layer-based Unified Canvas, node-based workflows, a model manager and a gallery. The hosted Invoke.ai platform has been shut down after the founding team joined Adobe, and the open-source project continues under community maintainers Lincoln Stein and Vic (Blessedcoolant).
What It Is
Invoke is a locally hosted web server with a React UI for generating, refining and iterating on images. Artists and enthusiasts can use it for text-to-image generation, in/out-painting, and editing on a layered canvas. Generated images are stored in boards and a gallery, with metadata that lets users recall the prompts and settings used.
Canvas, Workflows and Models
The Unified Canvas supports brush tools, in/out-painting and independently manipulated layers. Node-based workflows let users build reproducible generation pipelines as visual graphs and expose custom parameters in the UI so others can adjust values without editing the graph. The Model Manager handles checkpoints, LoRAs, Textual Inversions and ControlNets. The README lists support for SD 1.5, SDXL, SD 3.5, the Flux family, Qwen Image, Z-Image and others, plus API-only access to models such as GPT Image and Nano Banana. It also lists upscaling tools and object segmentation with SAM/SAM2.
Setup Path
The Invoke Launcher is the documented fastest way to install on Windows, macOS and Linux. Windows requires Windows 10 or later and an NVIDIA or AMD GPU. macOS requires Apple Silicon. Linux requires an NVIDIA or AMD GPU. Docker and manual Python installs are also documented, and hosted GPU options such as RunPod and Railway are linked. Invoke Training, a separate project, offers a UI for Textual Inversion and LoRA training.
Community Discussions
Be the first to start a conversation about InvokeAI
Share your experience with InvokeAI, ask questions, or help others learn from your insights.
Pricing
Community Edition (Open Source)
Free, self-hosted, Apache 2.0 licensed Invoke creative engine for AI image generation.
- Apache 2.0 licensed, self-hosted
- Unified Canvas with layer-based editing
- Node-based workflows
- Model Manager (Flux, SDXL, SD 1.5 and more)
- Runs locally on Windows, macOS, Linux
Capabilities
Key Features
- Layer-based Unified Canvas
- In/out-painting and brush tools
- Node-based workflows
- Model Manager for checkpoints, LoRAs, Textual Inversions and ControlNets
- ControlNet guidance
- Batch generation, prompt wildcards and upscaling
- Board and gallery management with image metadata recall
- Object segmentation with SAM/SAM2
- Local self-hosted operation
- Support for Flux, SDXL and SD 1.5 models
