Downloads · 30 days
0
c0ffeeOverdose/arn-manga-models
arn-manga-models is a machine learning model from c0ffeeOverdose. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as gpl-3.0.
Ready-to-download inference weights for the Arn Manga LLM Translator browser extension. This repo is a CDN bucket, not a model release — each file is a byte-identical conversion of upstream weights, mirrored here so t…
Downloads · 30 days
0
Access
Public
Updated Sep 16, 2026
Repo size
162 MB
Likes
0
Public
Click a slice to open those files.
.onnx162 MB · 100%
From the Hugging Face model README
Ready-to-download inference weights for the Arn Manga LLM Translator browser extension. This repo is a CDN bucket, not a model release — each file is a byte-identical conversion of upstream weights, mirrored here so the extension can fetch them on demand. If you train models, go star the originals.
| file | source | conversion | license |
|---|---|---|---|
ctd-int8.onnx (40MB, md5 ade48ae332291ef6cdbcd4257810c3ec) | lemondouble/lemon-manga-translator → onnx/comic-text-detector/ctd.onnx (upstream: dmMaze/comic-text-detector) | quantize_dynamic(WeightType.QUInt8), no pre-processing (see scripts/quantize-ctd.py) | GPL-3.0 |
lama-manga-512-fp16w.onnx (107MB, md5 8abe1d2e8e66d59d165e21cbf4a99a27) | dreMaz/AnimeMangaInpainting → lama_large_512px.ckpt (big-lama fine-tuned on 300k manga/anime pages) | fp16 weight-only + FFT-as-matmuls (ORT-web WebGPU has no DFT kernel; see scripts/export-inpaint-onnx.py) | MIT (model) / Apache-2.0 (architecture) |
panel-yolo26n.onnx (10MB, md5 21a14eb8206d44db688b575670023097) | leoxs22/manga-panel-detector-yolo26n → manga_panel_detector_fp32.pt (trained on Manga109-s) | YOLO(pt).export(format='onnx', imgsz=640, opset=17, dynamic=False, simplify=True) (see scripts/export-panel-onnx.sh) | Apache-2.0 (file) inside a GPL-3.0 collection |
.pt authors allow commercial use of results with dataset
attribution (Manga109-s, condition 5).