---

tags:

• gguf
• llama.cpp
• vision-language-model
• ocr
• latex
• mathematical-ocr

license: apache-2.0

---

Qwen3.5-4B LaTeX OCR (GGUF)

A GGUF version of the fine-tuned Qwen3.5-4B LaTeX OCR model for converting mathematical expressions from images into LaTeX.

Model Details

• Author: esalahterus
• Base Model: unsloth/Qwen3.5-4B
• Format: GGUF
• Framework: llama.cpp
• Task: Mathematical OCR (Image-to-LaTeX)
• License: Apache-2.0

Available Files

| File | Description |

|------|-------------|

| qwen3.5-4b-latex-ocr.Q8_0.gguf | Quantized GGUF model (Q8_0) |

| qwen3.5-4b-latex-ocr.F16-mmproj.gguf | Multimodal projection file (FP16) |

Usage

For multimodal inference with llama.cpp:

llama-mtmd-cli \
  -hf esalahterus/qwen3.5-4b-latex-ocr-gguf \
  --jinja

Description

This repository provides the GGUF version of the fine-tuned Qwen3.5-4B LaTeX OCR model, enabling efficient inference with llama.cpp and other GGUF-compatible runtimes.

Acknowledgements

This model was fine-tuned using Unsloth and converted to GGUF format for deployment with llama.cpp.

• Unsloth: https://github.com/unslothai/unsloth
• llama.cpp: https://github.com/ggml-org/llama.cpp