Downloads · 30 days
30
12% of all-time downloads
Qwen3.5-9B-AinuOCR
KoichiYasuoka/Qwen3.5-9B-AinuOCR
Qwen3.5-9B-AinuOCR is a image-text-to-text model from KoichiYasuoka. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
This is Qwen3.5-9B-based model, finetuned for OCR of Ainu texts written in カタカナ.
All-time downloads
252
Public
Parameters
9.4B
18.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors18.8 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- License
- apache-2.0
- Model type
- qwen3_5
- Access
- Public
- Created
- May 2, 2026
- Updated
- Oct 14, 2026
- SHA
- 3950106d
Base models
Model card
From the Hugging Face model README
Qwen3.5-9B-AinuOCR
Model Description
This is Qwen3.5-9B-based model, finetuned for OCR of Ainu texts written in カタカナ.
How to Use
from transformers import pipeline
img="https://huggingface.co/KoichiYasuoka/Qwen3.5-9B-AinuOCR/resolve/main/uwerankarap/orig/2026-01-31.jpg"
nlp=pipeline("image-text-to-text","KoichiYasuoka/Qwen3.5-9B-AinuOCR",max_new_tokens=2048)
doc=nlp([{"role":"user","content":[{"type":"image","image":img},{"type":"text","text":"OCR Ainu sentences."}]}])
print(doc[0]["generated_text"][1]["content"])
Reference
安岡孝一, 安岡素子: Qwen3.5によるカタカナアイヌ語OCRの開発, 情報処理学会研究報告, Vol.2026-CH-141『人文科学とコンピュータ』, No.3 (2026年5月16日), pp.1-5.
- Task
- Image-Text-to-Text
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- ain
- Created
- May 2, 2026
- Updated
- Oct 14, 2026