Metastable vs Stable Diffusion WebUI Forge
Metastable
metastable.studio
Stable Diffusion WebUI Forge
github.comSide by side
| Figure | Metastable | Stable Diffusion WebUI Forge |
|---|---|---|
| Price from | Free | Free |
| Free plan | Yes | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | — | Yes |
| Windows | Yes | Yes |
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| API | — | Yes |
| Self-hosted | — | Yes |
| Plans published | 1 | 1 |
| Spec score · AI image generators | 8.5 (#29) | 8.9 (#12) |
| Image editing | Yes | Yes |
| Installation | The app handles dependency installation, and its initial dependency download may be up to 5 GB. | The README recommends a one-click installation package that includes Git and Python, with CUDA 12.1 and PyTorch 2.3.1. |
| License | Metastable is open source and licensed under AGPLv3. | The repository identifies its license as AGPL-3.0. |
| Model support | The maker says Metastable supports Stable Diffusion 1, 2, 3, and XL models, FLUX.1 models, LoRAs, IP Adapters, and ControlNets. | The README says Flux BitsandBytes NF4 and several GGUF quantizations are natively supported, with LoRA support for specified NF4 and GGUF variants. |
| Purpose | Metastable is an app for AI image generation. | Forge is a platform built on Stable Diffusion WebUI and Gradio to ease development, optimize resource management, speed up inference, and study experimental features. |
| Reference images | Yes | Yes |
| Support | The documentation provides links to a Discord community and GitHub. | The maintainer invites users to open issues for problems and says they will check them every several days. |
| Text-to-image | Yes | Yes |
Plans
In their own words
Metastable
Metastable has no write-up on Specifiction yet; its figures are above.
Stable Diffusion WebUI Forge
Stable Diffusion WebUI Forge is a free, open-source platform built on Stable Diffusion WebUI for image generation and editing. It supports text-to-image creation, reference images, and image editing, with PNG, JPG, JPEG, WebP, AVIF, and GIF listed as output formats. The project aims to ease development, manage resources, speed inference, and explore experimental features. Its interface is based on Gradio, and the README offers a one-click installation package that includes Git and Python. Windows launch scripts, shell scripts, a macOS environment script, and source installation instructions are listed. Forge includes GPU memory management and components such as LoRAs, ControlNets, IP-Adapters, Instant-IDs, reference-only methods, and extensions. Several Flux BitsandBytes and GGUF variants are natively supported, with GPU weight and swap controls; some variants support LoRA. The README lists txt2img and img2img API endpoints as normal, while improved Flux support is pending. The project is described as under construction, so its documentation, interface, and functionality may change. Its license is AGPL-3.0.