bmaltais/kohya_ss (Gradio GUI for sd-scripts)
Source: https://github.com/bmaltais/kohya_ss
Structured fields
{
"archived": false,
"checked_at": "2026-10-17T00:00:00.000Z",
"parent_url": "https://github.com/kohya-ss/sd-scripts",
"usefulness": "Gradio GUI wrapper around kohya-ss/sd-scripts, the entry point most beginners use to configure LoRA and fine-tune runs. A wrapper, not the original trainer.",
"source_type": "repository",
"availability": "available",
"dependencies": [
"kohya-ss/sd-scripts (submodule/dependency)",
"Gradio",
"uv or pip install; Python 3.10 may be needed to avoid a tkinter error"
],
"next_check_at": "2026-10-24T00:00:00.000Z",
"hardware_notes": "Needs a GPU-equipped PC for local use; no VRAM figures on the page.",
"known_limitations": [
"GUI may lag the sd-scripts feature set; verify the pinned sd-scripts version per release."
],
"upstream_updated_at": "2026-08-01T05:43:00.000Z"
}Evidence claims
- [upstream_claim] README describes a Gradio-based GUI for Kohya's Stable Diffusion training scripts; local installs on Windows, Linux and macOS (macOS compatibility may vary) plus Colab/Runpod/Novita/Docker options. source
- [sourced_fact] Change history lists v26.0.0 as an upcoming release; the page does not confirm it has been published. source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | Apache-2.0 | permitted | permitted | permitted | unknown | Page states Apache License 2.0 (LICENSE.md). Bundled sd-scripts carries its own notices. |
| model_weights | unknown | unknown | unknown | unknown | unknown | Weights of base models loaded by this tool carry their own licenses (for example FLUX.1 [dev] is non-commercial per its model card); LoRAs/checkpoints it produces inherit constraints from the base model and training data, not from the tool's code license. Not assessed here. |