Project ◆ Hidden gem

Omost

lllyasviel/Omost · Python · Apache-2.0

LLM-composed canvases for structured concept art

Omost lets an LLM plan a canvas then fill regions — useful when game concept art needs layout control, not just a single prompt.

7.8 / 100 · Score
★ 7.6k Stars
+0 30-day growth
25 min Time to first run
Not required API key

Score

Trend 18
Activity 8
Quality 65
Community 27
Relevance 100

Setup at a glance

DeployWeb
GPU8 GB
API keyNot required
UI languagesen
Doc languagesen
Install
git clone https://github.com/lllyasviel/Omost.git && pip install -r requirements.txt

Deep dive

Omost is lllyasviel's compositional image generator: a coding LLM writes code that draws your image on a Canvas — every element placed, sized and layered explicitly instead of hoping a diffusion model guesses the layout.

That code-first approach makes it unusually good at structured scenes — multiple characters, precise object placement, UI-like compositions — which is exactly where pure prompt-based generators struggle.

It runs as a Hugging Face Space and a local Gradio app, with demo videos in the repo showing the code-to-image loop in action.

Use cases

Structured key art
Multi-region scenes
Concept boards

Good for / Not ideal for

Good for

Art directors and tech artists

Not ideal for

Users wanting pure pixel sprites

Trend · 30d

Alternatives