Instructions to use skumar9/Llama-medx_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use skumar9/Llama-medx_v2 with Transformers:

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

pipe = pipeline("text-generation", model="skumar9/Llama-medx_v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("skumar9/Llama-medx_v2")
model = AutoModelForCausalLM.from_pretrained("skumar9/Llama-medx_v2")
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 skumar9/Llama-medx_v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "skumar9/Llama-medx_v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "skumar9/Llama-medx_v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/skumar9/Llama-medx_v2

SGLang

How to use skumar9/Llama-medx_v2 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 "skumar9/Llama-medx_v2" \
    --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": "skumar9/Llama-medx_v2",
		"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 "skumar9/Llama-medx_v2" \
        --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": "skumar9/Llama-medx_v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use skumar9/Llama-medx_v2 with Docker Model Runner:
```
docker model run hf.co/skumar9/Llama-medx_v2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for Model ID

This is llama3 8b family chat model finetuned from base epfl-llm/meditron-7b with open assist dataset using SFT QLora .
All the linear parameters were made trainable with a rank of 16.

Prompt template: Llama

'<s> [INST] <<SYS>>
You are a helpful, respectful and medical honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.
If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
<</SYS>> {question} [/INST] {Model answer } </s>'

Usage:

model_name='jiviadmin/meditron-7b-guanaco-chat'

# Load the model
base_model = AutoModelForCausalLM.from_pretrained(
model_name,
low_cpu_mem_usage=True,
return_dict=True,
torch_dtype=torch.float16,
device_map={"": 0},
)
# Load tokenizer to save it
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True,add_eos_token=True)
tokenizer.add_special_tokens({'pad_token': '[PAD]'})
tokenizer.pad_token_id = 18610
tokenizer.padding_side = "right"

default_system_prompt="You are a helpful, respectful and honest medical assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.
If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.Please consider the context below if applicable:
Context:NA"

#Initialize the hugging face pipeline
def format_prompt(question):
  return f'''<s> [INST] <<SYS>> {default_system_prompt} <</SYS>> [INST] {question} [/INST]'''

question=' My father has a big white colour patch inside of his right cheek. please suggest a reason.'

pipe = pipeline(task="text-generation", model=base_model, tokenizer=tokenizer, max_length=512,repetition_penalty=1.1,return_full_text=False)
result = pipe(format_prompt(question))
answer=result[0]['generated_text']
print(answer)

Downloads last month: 11

Safetensors

Model size

8B params

Tensor type

F16

Model tree for skumar9/Llama-medx_v2

Quantizations

3 models

Paper for skumar9/Llama-medx_v2

QLoRA: Efficient Finetuning of Quantized LLMs

Paper • 2305.14314 • Published May 23, 2023 • 61