ko-hand-ocr-vl needs about 1.3 GB of memory at Q4_K_M (a 0.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.4 GB (Q4_K_M) to 0.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.
ko-hand-ocr-vl: Korean handwritten OCR VL model
ko-hand-ocr-vl is a 0.559B Qwen3.5-0.8B fine‑tuned for Korean handwritten OCR. It accepts full images and outputs text. The model is open‑weight and runs on Ollama, LM Studio, llama.cpp, vLLM, and transformers without extra code.
Good at
- Full‑image handwritten OCR
- Fast inference on 8‑GB GPUs
- High accuracy on unseen handwriting
Limits
- No support for languages other than Korean
- No additional modalities beyond vision
- License is apache‑2.0
Best for
- OCR of handwritten Korean documents
- Batch processing of scanned pages
- Integration into Korean language workflows
Who it’s for. Developers needing a lightweight Korean OCR solution for images. Researchers testing vision‑language models on handwritten data. Enterprises deploying OCR in Korean environments.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.4 GB | 1.3 GB | Fits | Get file → | |
| Q8_0 | 0.6 GB | 1.5 GB | Fits | Get file → |
For 8‑GB VRAM, use the Q4_K_M quantization (file size 0.36 GB, estimated 1.3 GB memory at 8k context). For larger GPUs, Q8_0 (0.6 GB, 1.5 GB memory) offers higher precision. Memory estimates are approximate.
What GPU do you need for ko-hand-ocr-vl?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q8_0 (0.6 GB file, about 1.5 GB in use).
Which ko-hand-ocr-vl quant should you choose?
Q4_K_M is the usual starting point for ko-hand-ocr-vl: it balances file size and quality well. Here it is a 0.4 GB file.
The Q8_0 file is 0.6 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about ko-hand-ocr-vl
How much VRAM does ko-hand-ocr-vl need?
About 1.3 GB at Q4_K_M with an 8K context window. That is the 0.4 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run ko-hand-ocr-vl on an 8 GB GPU?
Yes, with Q8_0: the 0.6 GB file needs about 1.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of ko-hand-ocr-vl GGUF?
It depends on the quant: Q4_K_M is 0.4 GB, Q8_0 is 0.6 GB.
What context length does ko-hand-ocr-vl support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.
Where can I download ko-hand-ocr-vl GGUF?
The files are on Hugging Face in the localdeel/ko-hand-ocr-vl-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does ko-hand-ocr-vl use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, the Q4_K_M quantized file fits 8 GB VRAM and uses about 1.3 GB of memory at 8k context.
Does it support languages other than Korean?
No, the model is trained only on Korean characters and tokens.
Is the model vision‑capable?
Yes, it reads full images and outputs the recognized text.