maqa_llama_4bit

MAQA-LLaMA-4bit — Arabic Medical Q&A (4-bit GPU Inference)

<p align="center"> <img src="https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png" width="115"/> </p>

⚠️ Disclaimer: This model is intended for research and informational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. It cannot and should not be used to prescribe or recommend medications.

---

Model Summary

maqa_llama_4bit is the 4-bit GPU inference variant of the MAQA-LLaMA family — a Llama 3 8B model fine-tuned on 430,000 real Arabic doctor-patient interactions across 20 medical specialisations. Runs on consumer-grade GPUs with 6–8 GB VRAM.

---

Model Family

---

Quick Start

With Unsloth (recommended — 2x faster inference)

from unsloth import FastLanguageModel
from unsloth.chat_templates import get_chat_template
from transformers import TextStreamer

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "AliAbdelrasheed/maqa_llama_4bit",
    max_seq_length = 2048,
    dtype = None,         # Auto-detect: Float16 for T4/V100, BFloat16 for Ampere+
    load_in_4bit = True,
)

tokenizer = get_chat_template(
    tokenizer,
    chat_template = "llama-3",
    mapping = {"role": "from", "content": "value", "user": "human", "assistant": "gpt"},
)

FastLanguageModel.for_inference(model)

messages = [
    {
        "from": "system",
        "value": "أنت طبيب محترف ولديك خبرة في كل مجالات الطب. يجيب على أسئلة المرضى حول الأمراض، باستخدام لهجة رسمية وودية، وإجابات موجزة ومفيدة يسهل على الجميع فهمها."
    },
    {
        "from": "human",
        "value": "أشعر بألم أسفل البطن بخاصرتي والألم يجي على فترات، ما الأسباب؟"
    },
]

inputs = tokenizer.apply_chat_template(
    messages,
    tokenize = True,
    add_generation_prompt = True,
    return_tensors = "pt",
).to("cuda")

streamer = TextStreamer(tokenizer)
_ = model.generate(
    input_ids = inputs,
    streamer = streamer,
    max_new_tokens = 256,
    use_cache = True,
)

With Transformers + bitsandbytes (no Unsloth required)

from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("AliAbdelrasheed/maqa_llama_4bit")
model = AutoModelForCausalLM.from_pretrained(
    "AliAbdelrasheed/maqa_llama_4bit",
    load_in_4bit = True,
    device_map = "auto",
)

prompt = "ما هي أعراض مرض السكري من النوع الثاني وكيف يمكن التعامل معه؟"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

---

Dataset — MAQA

Dataset: "Deep learning for Arabic healthcare: MedicalBot" — Springer (2023) Harvard Dataverse

---

Training Details

LoRA Configuration

Training Arguments

Full training details and notebook available on GitHub.

---

Limitations

• Not a substitute for professional medical advice or clinical diagnosis
• Cannot prescribe or recommend medications
• Trained on a sampled subset of MAQA; performance may vary across all 20 specialisations
• Optimised for Modern Standard Arabic; dialectal Arabic performance may vary
• Web-scraped data may contain noise or outdated medical information

---

Developed By

Ali Abdelrasheed — Graduation Project Nile University · B.Sc. Information Technology – Big Data · Class of 2024 🤗 HuggingFace Profile