A web app that generates 8-direction character turnaround sheets from a single uploaded image using Qwen-Image-Edit and multi-angle LoRAs, with per-view refinement, background removal, and automatic splitting of multi-character sheets.

Stars

29

7-day growth

No data

Forks

4

Open issues

0

License

MIT

Last updated

2026-07-09

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It combines state‑of‑the‑art I2I editing with a specialized multi‑angle LoRA for consistent character views, offers an automated sheet‑split feature for multi‑character input, and includes intelligent VRAM management with opt‑in reuse of ComfyUI model files.

Who it is for

  • Game developers needing consistent character turnarounds
  • Character designers and concept artists
  • 3D modelers who require multi‑view reference sheets
  • Hobbyists and indie creators using AI image generation

Use cases

  • Generating a standard 8‑view character sheet from a single illustration
  • Extracting individual characters from a crowded sheet and generating turnarounds for each
  • Refining a specific view (e.g., removing hair from the back) while keeping all other views intact
  • Batch downloading the full sheet and each view as a ZIP archive

Strengths

  • Automatically detects and splits multiple characters from a single uploaded sheet
  • Uses Lightning LoRA for fast 4‑step generation (~4‑5 sec per view on high‑end GPU)
  • Per‑view refinement with undo/redo toggle and automatic backup
  • VRAM adaptive offloading (4 levels) to run on GPUs as low as 24 GB

Considerations

  • Only one generation or refinement job can run at a time (concurrent requests return 409)
  • Job state is in‑memory and lost on server restart (output files remain on disk)
  • The included UI is in Japanese; international users must interact via the REST API

README quick start

使用モデル

Description

charsheet creation w/Diffusers

Related repositories

Similar projects matched by category, topics, and programming language.

lopopolo
Featured
lopopolo GitHub avatar

harness-engineering

Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.

AI & Machine LearningAI Agents
2,390
slvDev
Featured
slvDev GitHub avatar

esp32-ai

A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.

AI & Machine LearningLarge Language Models
1,960
littledivy
Featured
littledivy GitHub avatar

mimic

mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.

AI & Machine Learning
1,482