Instructions to use DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights with Transformers:

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

pipe = pipeline("text-generation", model="DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import MolLLaMA
model = MolLLaMA.from_pretrained("DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights

SGLang

How to use DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights 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 "DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights" \
    --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": "DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights",
		"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 "DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights" \
        --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": "DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights with Docker Model Runner:
```
docker model run hf.co/DongkiKim/Mol-Llama-3.1-8B-Instruct-Full-Weights
```

Mol-Llama-3.1-8B-Instruct-Full-Weights

[Project Page] [Paper] [GitHub]

This repo contains the weights of Mol-LLaMA including the LoRA weights and projectors, build with Llama: meta-llama/Llama-3.1-8B-Instruct. Llama 3.1 is licensed under the Llama 3.1 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.

Architecture

Molecular encoders: Pretrained 2D encoder (MoleculeSTM) and 3D encoder (Uni-Mol)
Blending Module: Combining complementary information from 2D and 3D encoders via cross-attention
Q-Former: Embed molecular representations into query tokens based on SciBERT
LoRA: Adapters for fine-tuning LLMs

Training Dataset

Mol-LLaMA is trained on Mol-LLaMA-Instruct, to learn the fundamental characteristics of molecules with the reasoning ability and explanbility.

Citation

If you find our model useful, please consider citing our work.

@misc{kim2025molllama,
    title={Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model},
    author={Dongki Kim and Wonbin Lee and Sung Ju Hwang},
    year={2025},
    eprint={2502.13449},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}