TostUI

Use it when

• You want to run a supported image-generation or editing model such as Flux.2 or Qwen Image Edit behind a local web UI without assembling its Python environment manually.
• You need one of the documented song, image-to-video, upscaling, deblurring, or 3D-generation interfaces on a compatible NVIDIA GPU, optionally on RunPod.

What it solves

Not the fit when

• CPU-only systems or machines without a compatible NVIDIA GPU; documented templates require between 6 GB and 24 GB VRAM.
• Headless API-driven batch pipelines; the documented containers expose browser interfaces.
• Assuming the repository's MIT license covers model weights or generated-content rights; verify each selected model's license and acceptable-use terms separately.
• Treating one Docker image as a universal interface: each model/template has a separate image, tag, resource requirement, and command.
• cloud API image generation
• cpu-only inference
• programmatic batch generation from code
• text LLM chat serving

Install

docker pull camenduru/tostui-flux-2-dev (one image per model template; requires Docker and an NVIDIA GPU)

Invoke

docker run --gpus all -p 3000:3000 --name tostui-flux-2-dev camenduru/tostui-flux-2-dev, then open http://localhost:3000 in a browser

Alternatives

No reviewed alternatives recorded yet.