Instructions to use Xianjun/Quokka-7b-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Xianjun/Quokka-7b-instruct with Transformers:

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

pipe = pipeline("text-generation", model="Xianjun/Quokka-7b-instruct")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Xianjun/Quokka-7b-instruct")
model = AutoModelForCausalLM.from_pretrained("Xianjun/Quokka-7b-instruct", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Xianjun/Quokka-7b-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Xianjun/Quokka-7b-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Xianjun/Quokka-7b-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Xianjun/Quokka-7b-instruct

SGLang

How to use Xianjun/Quokka-7b-instruct 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 "Xianjun/Quokka-7b-instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Xianjun/Quokka-7b-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Xianjun/Quokka-7b-instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Xianjun/Quokka-7b-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Xianjun/Quokka-7b-instruct with Docker Model Runner:
```
docker model run hf.co/Xianjun/Quokka-7b-instruct
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for Model ID

This model is optimized for Material Science by continuing pertaining on over 1 million Material science academic articles based on LLaMa-2-7b. And further finetuned on materials science instructions.

Developed by: [UCSB]
Language(s) (NLP): [More Information Needed]
License: [More Information Needed]
Finetuned from model [optional]: [LLaMa-2]
Paper [optional]: [https://arxiv.org/pdf/2401.01089.pdf]
Demo [optional]: [More Information Needed]

How to Get Started with the Model

from transformers import LlamaTokenizer, LlamaForCausalLM
import torch

tokenizer = LlamaTokenizer.from_pretrained("Xianjun/Quokka-7b-instruct ")
model = LlamaForCausalLM.from_pretrained("Xianjun/Quokka-7b-instruct ").half().to("cuda")

instruction = "How to ..."
batch = tokenizer(instruction, return_tensors="pt", add_special_tokens=False).to("cuda")
with torch.no_grad():
    output = model.generate(**batch, max_new_tokens=512, temperature=0.7, do_sample=True)
    response = tokenizer.decode(output[0], skip_special_tokens=True)

Citation

If you find Quokka useful in your research, please cite the following paper:

@inproceedings{Yang2024QuokkaAO,
  title={Quokka: An Open-source Large Language Model ChatBot for Material Science},
  author={Xianjun Yang and Stephen Wilson and Linda Ruth Petzold},
  year={2024},
  url={https://api.semanticscholar.org/CorpusID:266725577}
}

Downloads last month: 39

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for Xianjun/Quokka-7b-instruct

Quantizations

2 models

Paper for Xianjun/Quokka-7b-instruct

Quokka: An Open-source Large Language Model ChatBot for Material Science

Paper • 2401.01089 • Published Jan 2, 2024