Downloads · 30 days
19
22% of all-time downloads
qwen3.5-4b-latex-ocr
esalahterus/qwen3.5-4b-latex-ocr
qwen3.5-4b-latex-ocr is a image-to-text model from esalahterus. Use it when you need a caption or text from an image. It is set up for transformers. The card lists the license as apache-2.0.
A fine-tuned Qwen3.5-4B model for converting mathematical expressions from images into LaTeX.
All-time downloads
87
Public
Parameters
4.7B
9.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors9.3 GB · 100%
Parameter types
How the weights are stored.
BF164.7B · 100%
Base models
Model card
From the Hugging Face model README
Qwen3.5-4B LaTeX OCR
A fine-tuned Qwen3.5-4B model for converting mathematical expressions from images into LaTeX.
Model Details
- Author: esalahterus
- Base Model: unsloth/Qwen3.5-4B
- Framework: Unsloth + Hugging Face TRL
- Task: Mathematical OCR (Image to LaTeX)
- License: Apache-2.0
Description
This model is fine-tuned to recognize mathematical expressions from images and generate their corresponding LaTeX representations. It is intended for OCR tasks involving mathematical formulas and scientific documents.
Intended Use
- Mathematical formula recognition
- Image-to-LaTeX conversion
- Educational and research applications
- Document digitization
Base Model
This model is based on:
- unsloth/Qwen3.5-4B
Training
The model was fine-tuned using Unsloth and the Hugging Face TRL library on a custom mathematical OCR dataset.
License
Apache-2.0
- Task
- Image-to-Text
- Library
- transformers
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- en
- Created
- Jul 25, 2026
- Updated
- Jul 26, 2026