Instructions to use alpha-ai/LLAMA3-3B-Medical-COT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alpha-ai/LLAMA3-3B-Medical-COT with Transformers:

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

pipe = pipeline("text-generation", model="alpha-ai/LLAMA3-3B-Medical-COT")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("alpha-ai/LLAMA3-3B-Medical-COT")
model = AutoModelForCausalLM.from_pretrained("alpha-ai/LLAMA3-3B-Medical-COT")
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]:]))

Inference
Local Apps Settings

vLLM

How to use alpha-ai/LLAMA3-3B-Medical-COT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alpha-ai/LLAMA3-3B-Medical-COT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "alpha-ai/LLAMA3-3B-Medical-COT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/alpha-ai/LLAMA3-3B-Medical-COT

SGLang

How to use alpha-ai/LLAMA3-3B-Medical-COT 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 "alpha-ai/LLAMA3-3B-Medical-COT" \
    --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": "alpha-ai/LLAMA3-3B-Medical-COT",
		"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 "alpha-ai/LLAMA3-3B-Medical-COT" \
        --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": "alpha-ai/LLAMA3-3B-Medical-COT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use alpha-ai/LLAMA3-3B-Medical-COT with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for alpha-ai/LLAMA3-3B-Medical-COT to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for alpha-ai/LLAMA3-3B-Medical-COT to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for alpha-ai/LLAMA3-3B-Medical-COT to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="alpha-ai/LLAMA3-3B-Medical-COT",
    max_seq_length=2048,
)

Docker Model Runner
How to use alpha-ai/LLAMA3-3B-Medical-COT with Docker Model Runner:
```
docker model run hf.co/alpha-ai/LLAMA3-3B-Medical-COT
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Website - https://www.alphaai.biz

Uploaded Model - LLAMA3-3B-Medical-COT

Developed by: Alpha AI
License: Apache-2.0
Fine-tuned from model: LLAMA-3.2-3B-Instruct
This LLAMA-3.2-3B-Instruct model was fine-tuned using Unsloth and Hugging Face’s TRL library, ensuring efficient training and high-quality inference.

Newer versions

Overview

LLAMA3-3B-Medical-COT is a fine-tuned reasoning and medical problem-solving model built over LLAMA-3.2-3B-Instruct. The model is trained on a dataset focused on open-ended medical problems, aimed at enhancing clinical reasoning and structured problem-solving in AI systems.

This dataset consists of challenging medical exam-style questions with verifiable answers, ensuring factual consistency in responses. The fine-tuning process has strengthened the model’s chain-of-thought (CoT) reasoning, allowing it to break down complex medical queries step by step while maintaining conversational fluency.

Designed for on-device and local inference, the model is optimized for quick and structured reasoning, making it highly efficient for healthcare applications, academic research, and AI-driven medical support tools.

Model Details

Model: LLAMA-3.2-3B-Instruct
Fine-tuned By: Alpha AI
Training Framework: Unsloth + Hugging Face TRL
License: Apache-2.0
Format: GGUF (Optimized for local use)

Quantization Levels Available:

q4_k_m
q5_k_m
q8_0
16-bit Precision (This) Others - https://huggingface.co/alphaaico/LLAMA3-3B-Medical-COT-GGUF

Use Cases:

Medical Reasoning & Diagnosis Support – Assists in clinical discussions, case reviews, and problem-solving for medical professionals.
AI-Assisted Medical Learning – Enhances student learning through structured explanations and reasoning on medical exam questions.
Logical & Step-by-Step Problem Solving – Handles structured inference tasks beyond medical reasoning, making it useful in scientific research.
Conversational AI for Healthcare – Powers virtual assistants and AI-driven consultation tools with evidence-based responses.

Model Performance:

Fine-tuned on Verified Medical Reasoning Data – Ensures step-by-step logical responses grounded in medical accuracy.
Optimized for Local Deployment – Runs efficiently on personal GPUs and edge devices without requiring cloud infrastructure.
Structured Thought Process – Breaks down complex medical questions into logical, evidence-based answers.

Limitations & Biases:

While trained on verified medical datasets, this model is not a replacement for professional medical advice and should be used as a supplementary tool rather than a definitive diagnostic system. The model may exhibit biases from its dataset, and responses should always be validated by medical experts before being used in real-world applications.

Acknowledgments Special thanks to:

Unsloth for optimizing fine-tuning pipelines.
Hugging Face TRL for robust model training tools.
Dataset contributors for providing structured medical reasoning problems.

Downloads last month: 1,398

Model tree for alpha-ai/LLAMA3-3B-Medical-COT

Quantizations

2 models

Collection including alpha-ai/LLAMA3-3B-Medical-COT

Finetunes | SLMs and LLMs

Collection

Various variants of LLMs finetuned using proprietary data. • 27 items • Updated Jan 19 • 4