LLM EXPLORER 60,645 MODELS INDEXED

Vitstr Small Korean by Bingsu

By Bingsu · 11 downloads
HF Repo ↗ Copy URL

Vitstr Small Korean is an open-source language model by Bingsu. Features: 21.4m LLM, VRAM: 0.1GB, License: mit, LLM Explorer Score: 0.05.

  En   Endpoints compatible   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/Bingsu/vitstr_small-korean 

Vitstr Small Korean Parameters and Internals

Model Type 
Optical Character Recognition
Use Cases 
Areas:
Text recognition in images
Applications:
Document digitization, text extraction
Primary Use Cases:
Optical character recognition across various domains
Additional Notes 
Doctr OCR model handles recognition and detection tasks separately, leveraging pre-trained architectures for optimal performance.
Supported Languages 
Korean (Proficient)
Training Details 
Data Sources:
Custom OCR training dataset
Data Volume:
1000 training samples, 20 validation samples
Methodology:
OCR recognition using vitstr_small architecture
Model Architecture:
vitstr_small
Input Output 
Input Format:
Images
Accepted Modalities:
Image
Output Format:
Recognized text
LLM NameVitstr Small Korean
Repository 🤗https://huggingface.co/Bingsu/vitstr_small-korean 
Model Size21.4m
Required VRAM0.1 GB
Updated2026-09-07
MaintainerBingsu
Model Files  0.1 GB   0.1 GB
Supported Languagesen
Model ArchitectureAutoModel
Licensemit

Best Alternatives to Vitstr Small Korean

Best Alternatives
Context / RAM
Downloads
Likes
...uxia 21.4B Alignment V1.0 GGUF0K / 8.1 GB4609
Luxia 21.4B Q4 K M GGUF0K / 12.9 GB420
Luxia 21.4B 16bit GGUF0K / 42.8 GB150
Note: green Score (e.g. "73.2") means that the model is better than Bingsu/vitstr_small-korean.