Instructions to use mjpsm/Llama-3.2-1B-Instruction-Chatbot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mjpsm/Llama-3.2-1B-Instruction-Chatbot with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="mjpsm/Llama-3.2-1B-Instruction-Chatbot")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mjpsm/Llama-3.2-1B-Instruction-Chatbot")
model = AutoModelForCausalLM.from_pretrained("mjpsm/Llama-3.2-1B-Instruction-Chatbot")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mjpsm/Llama-3.2-1B-Instruction-Chatbot with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mjpsm/Llama-3.2-1B-Instruction-Chatbot"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mjpsm/Llama-3.2-1B-Instruction-Chatbot",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mjpsm/Llama-3.2-1B-Instruction-Chatbot

SGLang

How to use mjpsm/Llama-3.2-1B-Instruction-Chatbot with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "mjpsm/Llama-3.2-1B-Instruction-Chatbot" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mjpsm/Llama-3.2-1B-Instruction-Chatbot",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "mjpsm/Llama-3.2-1B-Instruction-Chatbot" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mjpsm/Llama-3.2-1B-Instruction-Chatbot",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mjpsm/Llama-3.2-1B-Instruction-Chatbot with Docker Model Runner:
```
docker model run hf.co/mjpsm/Llama-3.2-1B-Instruction-Chatbot
```

Llama 3.2 1B Instruction Chatbot

Overview

This repository contains a fully merged version of a LoRA fine-tuned Llama 3.2 1B Instruct model.

Unlike the adapter repository, this model can be loaded directly with the Hugging Face Transformers library without requiring PEFT.

The model was created to demonstrate the complete lifecycle of modern large language model fine-tuning, including:

Dataset preparation
Chat formatting
QLoRA training
LoRA merging
Hugging Face deployment

Base Model

meta-llama/Llama-3.2-1B-Instruct

Training Dataset

Approximately 40,000 instruction-response examples formatted into conversational chat format.

Training Configuration

Training Method

QLoRA

Merged Into

Full standalone Transformers model

Frameworks

Transformers
TRL
PEFT
Accelerate
BitsAndBytes

Hyperparameters

Parameter	Value
Epochs	1
Batch Size	2
Gradient Accumulation	4
Learning Rate	2e-4
Optimizer	paged_adamw_8bit
Scheduler	Cosine
Precision	FP16

Training Results

Metric	Value
Training Loss	0.15088
Runtime	6789.63 seconds
Global Steps	5000
Samples / Second	5.89
Steps / Second	0.736

Intended Uses

Conversational AI
Instruction following
Educational projects
Fine-tuning demonstrations
Research

Limitations

This model has not been benchmarked against established instruction-following datasets such as MT-Bench, AlpacaEval, or MMLU.

It has not undergone additional safety alignment beyond the original base model.

Responses may contain hallucinations or factual inaccuracies.

Loading the Model

from transformers import AutoTokenizer
from transformers import AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained(
    "mjpsm/Llama-3.2-1B-Instruction-Chatbot"
)

tokenizer = AutoTokenizer.from_pretrained(
    "mjpsm/Llama-3.2-1B-Instruction-Chatbot"
)

Citation

If you use this model in your work, please cite this repository.

Downloads last month: 61

Safetensors

Model size

1B params

Tensor type

F16