feat: anime-text model

fix: ZLUDA compatibility issues (#582)

feat: Flux.2 Klein inpainting

fix: inpainting OOM & pipeline error handling

refactor(ml): shared TextRegion/TextDirection types for pipeline engines

feat: add AMD GPU support via ZLUDA (#362)

feat: blob API, bubble/AOT-inpainting models, and rendering redesign

feat: custom tracing layer and log suppression

feat: pluggable engine system with DAG pipeline

feat: bootstrap window

refactor: merge crates

feat: integrate llama.cpp

feat: manga-text-segmentation-2025 model

feat: paddleocr-vl-1.5 model

feat: pp-doclayout-v3 model

feat: support more languages

feat: add mit48px OCR backend

refactor: reduce duplicates

refactor: remove sys-locale

refactor: simplify device name logic

refactor: split modules

feat: download progress indicator

feat: device info

feat: multi lang llm translate

feat: add hunyuan to translate more languages

feat: font detection using yuzumaker.FontDetection

tests: add ctd integration test

fix: support non-nvidia gpu on windows

refactor: re-introduce koharu-core

fix: hf hub download also uses cache location

chore: format code

fix: portable mode libs and models location

feat: add hf mirror

refactor: use koharu-ml

refactor: rename models to ml