Downloads · 30 days
26
10% of all-time downloads
naazimsnh02/medical-ministral-8b
medical-ministral-8b is a text generation model from naazimsnh02. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
<p align="center" <img src="./cover.jpg" width="850" </p
Downloads · 30 days
26
10% of all-time downloads
All-time downloads
250
Public
Parameters
8.9B
41.9 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors24.1 GB · 100%
From the Hugging Face model README
A medical reasoning model fine-tuned on Ministral-3-8B-Reasoning-2512 for clinical question answering and diagnostic reasoning.
This model is a specialized medical AI assistant trained in two phases:
The model demonstrates strong reasoning capabilities for medical scenarios, particularly in differential diagnosis and clinical decision-making.
The model was evaluated on the MMLU-Pro benchmark in the Biology and Health domains to assess medical and biomedical reasoning capabilities.
| Domain | Samples Evaluated | Accuracy |
|---|---|---|
| Biology & Health | 270 | 68.9% |
| Domain | Samples Evaluated | Accuracy |
|---|---|---|
| Biology | 390 | 68.7% |
| Health | 390 | 56.1% |
pip install transformers==5.0.0rc0
pip install mistral-common --upgrade
import torch
from transformers import Mistral3ForConditionalGeneration, MistralCommonBackend
model_id = "naazimsnh02/medical-ministral-8b"
tokenizer = MistralCommonBackend.from_pretrained(model_id)
model = Mistral3ForConditionalGeneration.from_pretrained(
model_id,
torch_dtype=torch.float16,
device_map="auto",
)
def chat(prompt: str) -> str:
messages = [{"role": "user", "content": prompt}]
tokenized = tokenizer.apply_chat_template(
messages,
return_tensors="pt",
return_dict=True
)
device = model.device
for key, val in tokenized.items():
if torch.is_tensor(val):
tokenized[key] = val.to(device)
input_length = tokenized["input_ids"].shape[1]
output = model.generate(
**tokenized,
max_new_tokens=4096,
temperature=0.7,
do_sample=True,
top_p=0.9,
)[0]
generated_tokens = output[input_length:]
decoded = tokenizer.decode(
generated_tokens,
skip_special_tokens=True
)
return decoded.strip()
# Example
print(chat("Given the symptoms of sudden weakness in the left arm and leg, recent long-distance travel, and the presence of swollen and tender right lower leg, what specific cardiac abnormality is most likely to be found upon further evaluation that could explain these findings?"))
✅ Patient Education: Explaining medical conditions in accessible language
✅ Clinical Decision Support: Providing reasoning for differential diagnoses
✅ Medical Content Generation: Creating educational materials and summaries
✅ Healthcare Chatbots: Powering conversational medical assistance
✅ Medical Training: Supporting medical students with explanations
❌ Not a replacement for professional medical advice
❌ Not for emergency medical situations
❌ Not validated for direct clinical decision-making
❌ Not trained on patient-specific health records
⚠️ Important: This model is designed for educational and informational purposes. Always consult qualified healthcare professionals for medical decisions.
If you use this model in your research, please cite:
@misc{medical-ministral-8b,
author = {naazimsnh02},
title = {Medical Ministral 8B: A Clinical Reasoning Model},
year = {2025},
publisher = {HuggingFace},
howpublished = {\url{https://huggingface.co/naazimsnh02/medical-ministral-8b}}
}
This model inherits the license from the base Ministral model. Please refer to the base model's license for usage terms.
Disclaimer: This model is provided "as-is" for research and educational purposes. The creators are not liable for any consequences resulting from the use of this model. Always consult qualified healthcare professionals for medical advice.