hikari 5bfd25e60d
Security Scan and Upload / Security & DefectDojo Upload (push) Successful in 1m13s
CI / Lint & Check (push) Successful in 13m23s
CI / Build Windows (push) Successful in 36m51s
fix(ci): pin LLVM 18 for Windows cross-compilation (#17)
## Summary

The Windows build job has been failing on every PR since the initial release. The root cause is that the default `clang`/`lld`/`llvm` apt packages resolve to an older version of `llvm-rc` that cannot handle PNG-compressed entries in `.ico` files — and all six images in `icon.ico` use PNG compression.

Pinning to `clang-18`, `lld-18`, and `llvm-18` (available in Ubuntu 24.04's default repos) and registering them via `update-alternatives` ensures `llvm-rc`, `clang-cl`, `lld-link`, and friends all resolve to a version that handles the icon resource correctly.

 This PR was created with help from Hikari~ 🌸

Reviewed-on: #17
Co-authored-by: Hikari <hikari@nhcarrigan.com>
Co-committed-by: Hikari <hikari@nhcarrigan.com>
2026-04-13 16:21:42 -07:00
2026-04-09 20:29:46 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 20:22:49 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 15:11:20 -07:00
2026-04-09 20:16:54 -07:00
2026-04-09 20:16:54 -07:00

Tatsumi

Tatsumi is a desktop app for generating AI character art powered by Google Gemini's image model. It's built specifically to generate art of Naomi, with her reference sheet pre-loaded so you just describe the scene.

Features

Three generation modes

  • Avatar — Generate a portrait or avatar from a text prompt (1:1 aspect ratio)
  • Art — Generate a full widescreen piece from a text prompt (16:9 aspect ratio)
  • Replace — Upload a source image and describe what you want changed or reimagined

Conversation threads

Each generation session is a thread. You can send follow-up messages to refine the result, retry any response, edit a previous prompt and re-run, or delete messages. Threads are saved locally and persist between sessions.

Other bits

  • Model reasoning is shown as a collapsible "Model reasoning" section on each response
  • Cost per response is displayed (input + output tokens)
  • Generated images can be downloaded directly from the chat
  • App version shown in the sidebar header

Setup

You'll need a Google Gemini API key to use Tatsumi. On first launch, you'll be prompted to enter it. You can update it any time via Settings.

Running locally

pnpm install
pnpm tauri dev

Building

pnpm tauri build
# or for Windows cross-compilation:
pnpm build:windows

Feedback and Bugs

If you have feedback or a bug report, please log a ticket on our forum.

Contributing

If you would like to contribute to the project, you may create a Pull Request containing your proposed changes and we will review it as soon as we are able! Please review our contributing guidelines first.

Code of Conduct

Before interacting with our community, please read our Code of Conduct.

License

This software is licensed under our global software license.

Copyright held by Naomi Carrigan.

Contact

We may be contacted through our Chat Server or via email at contact@nhcarrigan.com.

S
Description
Desktop app for generating Naomi art~
Readme 42 MiB
v1.2.0 Latest
2026-04-28 14:58:08 -07:00
Languages
TypeScript 82.4%
Rust 16.1%
JavaScript 0.6%
CSS 0.6%
HTML 0.3%