Instructions to use TheAgenticAI/mistral-medical with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheAgenticAI/mistral-medical with Transformers:

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

pipe = pipeline("text-generation", model="TheAgenticAI/mistral-medical")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TheAgenticAI/mistral-medical")
model = AutoModelForCausalLM.from_pretrained("TheAgenticAI/mistral-medical")
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 TheAgenticAI/mistral-medical with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/TheAgenticAI/mistral-medical

SGLang

How to use TheAgenticAI/mistral-medical 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 "TheAgenticAI/mistral-medical" \
    --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": "TheAgenticAI/mistral-medical",
		"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 "TheAgenticAI/mistral-medical" \
        --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": "TheAgenticAI/mistral-medical",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TheAgenticAI/mistral-medical with Docker Model Runner:
```
docker model run hf.co/TheAgenticAI/mistral-medical
```

mistral-medical

This model was trained from scratch on an TheAgenticAI/medical-o1-reasoning-SFT dataset. It achieves the following results on the evaluation set:

Loss: 1.2238
Token Acc: 0.6780

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

TheAgenticAI/medical-o1-reasoning-SFT

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-06
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 16
total_train_batch_size: 128
total_eval_batch_size: 8
optimizer: Use adamw_torch with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.05
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Token Acc
1.0816	0.6841	100	1.0844	0.6876
0.8230	1.3625	200	1.1134	0.6851
0.5641	2.0410	300	1.1864	0.6809
0.5385	2.7251	400	1.2243	0.6781
0.4993	2.9850	438	1.2238	0.6780

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.2.0
Tokenizers 0.21.1

Downloads last month: 8

Safetensors

Model size

24B params

Tensor type

BF16