---
library_name: transformers
pipeline_tag: text-classification
tags:
  - deberta
  - multiple-choice
  - mcq
  - question-answering
---

# DeBERTa MCQ Solver

A DeBERTa-based multiple-choice question solver trained using PyTorch and Hugging Face Transformers.

## Model

This model is trained for multiple-choice question answering with five candidate options.

### Architecture

- Base model: DeBERTa
- Framework: PyTorch
- Library: Hugging Face Transformers
- Number of choices: 5
- Maximum sequence length: 256

## Files

- `config.json` — model configuration
- `model.safetensors` — trained model weights
- `tokenizer.json` — tokenizer
- `tokenizer_config.json` — tokenizer configuration
- `app.py` — Gradio application
- `requirements.txt` — Python dependencies

## Usage

```python
from transformers import AutoTokenizer, AutoModelForMultipleChoice

model_name = "rohitk123/deberta-mcq-solver"

tokenizer = AutoTokenizer.from_pretrained(model_name)

model = AutoModelForMultipleChoice.from_pretrained(model_name)