Instructions to use ik-ram28/MedMistral-CPT-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ik-ram28/MedMistral-CPT-7B with Transformers:

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

pipe = pipeline("text-generation", model="ik-ram28/MedMistral-CPT-7B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ik-ram28/MedMistral-CPT-7B")
model = AutoModelForCausalLM.from_pretrained("ik-ram28/MedMistral-CPT-7B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ik-ram28/MedMistral-CPT-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ik-ram28/MedMistral-CPT-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ik-ram28/MedMistral-CPT-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ik-ram28/MedMistral-CPT-7B

SGLang

How to use ik-ram28/MedMistral-CPT-7B 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 "ik-ram28/MedMistral-CPT-7B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ik-ram28/MedMistral-CPT-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "ik-ram28/MedMistral-CPT-7B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ik-ram28/MedMistral-CPT-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ik-ram28/MedMistral-CPT-7B with Docker Model Runner:
```
docker model run hf.co/ik-ram28/MedMistral-CPT-7B
```

Model Description

MedMistral-CPT-7B is a French medical language model based on Mistral-7B-v0.1, adapted for medical domain applications through Continual Pre-Training (CPT) on French medical texts.

Model Details

Model Type: Causal Language Model
Base Model: Mistral-7B-v0.1
Language: French
Domain: Medical/Healthcare
Parameters: 7 billion
License: Apache 2.0

Training Details

Continual Pre-Training (CPT)

Dataset: NACHOS corpus (7.4 GB French medical texts)
Training Duration: 2.8 epochs
Hardware: 32 NVIDIA H100 80GB GPUs
Training Time: ~40 hours

Computational Impact

Carbon Emissions: 9.86 kgCO2e
Training Time: 12 hours

Ethical Considerations

Medical Accuracy: For research and educational purposes only
Professional Oversight: Requires verification by qualified medical professionals
Bias Awareness: May contain biases from training data
Privacy: Do not input private health information

Citation

Contact

For questions about these models, please contact: ikram.belmadani@lis-lab.fr

Downloads last month: 4

Safetensors

Model size

7B params

Tensor type

F32

Model tree for ik-ram28/MedMistral-CPT-7B

Base model

mistralai/Mistral-7B-v0.1

Finetuned

(1033)

this model

Finetunes

1 model