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

Libraries

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

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

pipe = pipeline("text-generation", model="ik-ram28/MedMistralInstruct-CPT-SFT-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

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

vLLM

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

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ik-ram28/MedMistralInstruct-CPT-SFT-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ik-ram28/MedMistralInstruct-CPT-SFT-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

SGLang

How to use ik-ram28/MedMistralInstruct-CPT-SFT-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/MedMistralInstruct-CPT-SFT-7B" \
    --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": "ik-ram28/MedMistralInstruct-CPT-SFT-7B",
		"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 "ik-ram28/MedMistralInstruct-CPT-SFT-7B" \
        --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": "ik-ram28/MedMistralInstruct-CPT-SFT-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ik-ram28/MedMistralInstruct-CPT-SFT-7B with Docker Model Runner:
```
docker model run hf.co/ik-ram28/MedMistralInstruct-CPT-SFT-7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

MedMistralInstruct-CPT-SFT-7B

Model Description

MedMistralInstruct-CPT-SFT-7B is a French medical language model based on Mistral-7B-Instruct-v0.1, adapted through Continual Pre-Training followed by Supervised Fine-Tuning.

Model Details

Model Type: Causal Language Model
Base Model: Mistral-7B-Instruct-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 A100 80GB GPUs
Training Time: ~40 hours

Supervised Fine-Tuning (SFT)

Dataset: 30K French medical question-answer pairs
Method: DoRA (Weight-Decomposed Low-Rank Adaptation)
Training Duration: 10 epochs
Hardware: 1 NVIDIA H100 80GB GPU
Training Time: ~42 hours

Computational Requirements

Carbon Emissions: 33.96 kgCO2e (CPT+SFT)
Training Time: 82 hours total (CPT+SFT)

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