ML-powered manga translator, written in Rust.
Getting Started · Docs · Bug reports · Discord
Koharu introduces a local-first workflow for manga translation, utilizing the power of ML to automate the process. It combines the capabilities of object detection, OCR, inpainting, and LLMs to create a seamless translation experience.
Note
Koharu runs its vision models and LLMs locally on your machine to keep your data private and secure.
Note
Support and discussion are available on the Discord server.
- Automatic detection of text regions, speech bubbles, and cleanup masks
- OCR for manga dialogue, captions, and other page text
- Inpainting to remove source lettering from the page
- Translation with local or remote LLM backends
- Advanced text rendering with vertical CJK and RTL support
- Layered PSD export with editable text
Koharu supports CUDA, ROCm / HIP, Metal, and Vulkan. CPU fallback is always available when the accelerated path is unavailable or not worth the setup cost on your system.
Koharu supports NVIDIA GPUs on Windows and Linux through CUDA. Ensure you have the latest NVIDIA driver installed.
Koharu supports AMD GPUs on Windows through ROCm and HIP. Ensure you have the latest AMD driver installed.
Koharu supports Metal on Apple Silicon Macs.
Koharu also supports Vulkan on Windows and Linux as an alternative to CUDA and HIP.
You can always force Koharu to use CPU for inference:
# macOS / Linux
koharu --cpu
# Windows
koharu.exe --cpuKoharu uses a staged stack of vision and language models instead of trying to solve the entire page with a single network.
Koharu uses multiple pretrained models, each tuned for a specific part of the page pipeline.
Koharu uses object detection to find text regions, speech bubbles, and segmentation masks.
These models recognize source text after detection.
These models remove source lettering before translated text is rendered back onto the page.
Koharu has a flexible LLM backend that can run locally or connect to a remote API.
- LFM 2.5: lfm2.5-1.2b-instruct
- Ministral 3: ministral-3-8b-instruct
- Gemma 4 instruct (QAT): gemma4-e2b-it, gemma4-e4b-it, gemma4-12b-it, gemma4-26b-a4b-it, gemma4-31b-it
- Qwen 3.5: qwen3.5-0.8b, qwen3.5-2b, qwen3.5-4b, qwen3.5-9b, qwen3.5-27b, qwen3.5-35b-a3b
- Qwen 3.6: qwen3.6-27b, qwen3.6-35b-a3b
- Gemma 4 uncensored: gemma4-e2b-uncensored, gemma4-e4b-uncensored, gemma4-12b-uncensored, gemma4-26b-a4b-uncensored, gemma4-31b-uncensored
- Qwen 3.5 uncensored: qwen3.5-2b-uncensored, qwen3.5-4b-uncensored, qwen3.5-9b-uncensored
- Qwen 3.6 uncensored: qwen3.6-27b-uncensored, qwen3.6-35b-a3b-uncensored
Koharu supports hosted APIs from Atlas Cloud, OpenAI, Gemini, Claude, DeepSeek, and OpenRouter.
For pure machine-translation use cases, Koharu also supports DeepL, Google Cloud Translation, and Caiyun.
Koharu supports any provider that implements the OpenAI-compatible API.
You can download the latest release of Koharu from the releases page.
We provide prebuilt binaries for Windows, macOS, and Linux.
On Windows, you can install Koharu with winget:
winget install koharuOn macOS, you can install Koharu with Homebrew:
brew install --cask koharuKoharu provides a diagnostic mode that outputs detailed logs and system information to help identify issues with installation, GPU acceleration, model loading, and more. To enable it, run:
# macOS / Linux
koharu --debug
# Windows
koharu.exe --debugYou can also set the RUST_LOG environment variable to debug or trace to see more verbose logs:
# macOS / Linux
RUST_LOG=debug koharu
# Windows (PowerShell)
$env:RUST_LOG="debug"; koharu.exeTo build Koharu from source, follow the steps below.
bun installbun devbun run buildThe built binaries are written to target/release.
If Koharu is useful in your workflow, consider sponsoring the project.
Thanks to all the contributors who have helped make Koharu better!
Koharu is licensed under the GNU General Public License v3.0.