koharu — What is it?

Koharu is an open-source, ML-powered manga translation tool that automates the process of translating manga pages while preserving visual integrity.

⭐ 2,371 Stars 🍴 124 Forks Rust Author: mayocream
Source: README View on GitHub →

Why it matters

Koharu is gaining attention due to its innovative approach to manga translation, addressing the need for accurate and visually consistent translations. Its use of local ML models ensures privacy and performance, and its Rust-based architecture highlights its focus on safety and speed.

Source: Synthesis of README and project traits

Core Features

Automatic Text Detection and Cleanup

Koharu automatically detects text regions, speech bubbles, and cleanup masks, using object detection and OCR to ensure accurate translation without manual intervention.

Source: README
Inpainting and Translation

The tool employs inpainting to remove source lettering and translates text using local or remote LLM backends, providing a seamless translation experience.

Source: README
Advanced Text Rendering

Koharu supports vertical CJK and RTL text rendering, ensuring that translated text appears natural within speech bubbles and captions.

Source: README
Layered PSD Export

Users can export translated pages as layered PSD files, preserving helper layers and editable text layers for further refinement.

Source: README
Local HTTP API and MCP Server

Koharu includes a local HTTP API and MCP server for automation and integration with other tools.

Source: README

Architecture

Koharu's architecture is modular, with separate components for vision and language processing. It uses Rust for safety and performance, leveraging Tauri for the desktop app and candle and llama.cpp for ML inference. The project employs a staged stack of models for different stages of the translation process.

Source: Code tree + dependency files

Project Knowledge Graph

Knowledge graph: project (center) + core features (inner hexagons) + key dependencies (outer chips) candle llama.cpp Tauri candle-transformerscandle-transfo… candle-core Automatic Text Detection and CleanupAutomatic Text Dete… Inpainting and TranslationInpainting and Tran… Advanced Text RenderingAdvanced Text Rende… Layered PSD Export Local HTTP API and MCP ServerLocal HTTP API and… koharu Project Core feature Key dependency

Center: project; inner ring: core feature modules; outer ring: key dependencies. Auto-generated from core_features and tech_stack.key_deps.

Tech Stack

LanguageRustFrameworkTauri, candle, llama.cpp
candlellama.cppTauricandle-transformerscandle-corecandle-nncandle-flash-attnhf-hubimageanyhowasync-traitimageprocclapserdeserde_bytesserde_jsontokiotracingtracing-subscriberconsoleopenrfdpostcardthiserrortauritauri-build
Not enough information.
Source: Dependency files + code tree

Quick Start

To install Koharu, run the following command: ```bash koharu --port 9999 ```
Source: README Installation/Quick Start

Use Cases

Koharu is suitable for manga enthusiasts, translators, and developers looking to automate the translation of manga pages. It can be used for personal projects, educational purposes, or to create professional translations.

Source: README

Strengths & Limitations

Strengths

  • Strength 1: Local ML models ensure privacy and performance.
  • Strength 2: Modular architecture allows for easy updates and maintenance.
  • Strength 3: Supports a wide range of manga page layouts and text rendering.

Limitations

  • Limitation 1: The project is still under development and may have bugs or limitations.
  • Limitation 2: The project's performance may vary depending on the user's hardware.
Source: Synthesis of README, code structure and dependencies

Latest Release

Version 0.58.0, released on 2026-05-01. This release includes bug fixes and new features such as Chinese segmentation jieba and improved text rendering.

Source: GitHub Releases

Verdict

Koharu is a promising open-source project for manga translation, offering a unique combination of privacy, performance, and ease of use. It is particularly suitable for individuals and teams interested in exploring the intersection of machine learning and manga translation.

Source: Synthesis
Transparency Notice
This page is auto-generated by AI (a large language model) from the following public materials: GitHub README, code tree, dependency files and release notes. Analyzed at: 2026-05-23 19:21. Quality score: 85/100.

Data sources: README, GitHub API, dependency files