---
license: apache-2.0
language:
- en
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
- roast
- fun
- humor
- chatbot
- text-generation
library_name: transformers
---

# BurnMaster-1B 🔥

## Introduction

**BurnMaster-1B** is a playful AI roast-bot built on top of [Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).  
Instead of being a generic assistant, BurnMaster specializes in delivering **short, witty, clean roasts** for fun and entertainment.  

✨ Features:
- Clean & funny burns, safe for friends
- Adjustable *spice levels* (from teasing → max spicy roast)
- Preloaded with random roast ideas for instant laughs
- Powered by the Qwen2.5-1.5B-Instruct model architecture

## Quickstart

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "CoderPixel/BurnMaster-1B"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)

messages = [
    {"role": "system", "content": "You are BurnMaster, an AI that delivers short, funny, clean roasts."},
    {"role": "user", "content": "Roast me like I rage quit Roblox after losing to a 9-year-old."}
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer([text], return_tensors="pt").to(model.device)

outputs = model.generate(**inputs, max_new_tokens=100, do_sample=True, temperature=0.9, top_p=0.9)
response = tokenizer.decode(outputs[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)

print(response)