Instructions to use sagnik-p/medical_llm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sagnik-p/medical_llm with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sagnik-p/medical_llm")
model = AutoModelForCausalLM.from_pretrained("sagnik-p/medical_llm")
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 sagnik-p/medical_llm with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/sagnik-p/medical_llm

SGLang

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

Docker Model Runner
How to use sagnik-p/medical_llm with Docker Model Runner:
```
docker model run hf.co/sagnik-p/medical_llm
```

Medichat-V2-Llama3-8B

This is a merge of pre-trained language models created using mergekit.

This model is particularly effective in structuring the unstructured clinical texts.

Model Composition and Features:

Base Model: The foundation of this model is based on "refuelai/Llama-3-Refueled," which itself is a refined version of the Llama3-8B model, renowned for its instruction-following capabilities and adaptability across various domains.
Merged Models:
- ruslanmv/ai-medical-model-32bit: A model fine-tuned specifically for answering technical medical questions, providing a solid base of medical knowledge.
- Locutusque/Llama-3-Hercules-5.0-8B: Known for its ability to follow complex instructions and handle conversational interactions effectively, especially in scientific and technical contexts.

This model was merged using the DARE TIES merge method using refuelai/Llama-3-Refueled as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:



models:
  - model: Locutusque/Llama-3-Hercules-5.0-8B
    parameters:
      weight: [0.25, 0.45, 0.35, 0.45, 0.25]
      density: [0.1, 0.25, 0.5, 0.25, 0.1]
  - model: refuelai/Llama-3-Refueled
  - model: ruslanmv/ai-medical-model-32bit
    parameters:
      weight: [0.55, 0.45, 0.25, 0.45, 0.55]
      density: [0.1, 0.25, 0.5, 0.25, 0.1]
merge_method: dare_ties
base_model: refuelai/Llama-3-Refueled
parameters:
  int8_mask: true
dtype: bfloat16