Instructions to use Artigenz/Artigenz-Coder-DS-6.7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Artigenz/Artigenz-Coder-DS-6.7B with Transformers:

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

pipe = pipeline("text-generation", model="Artigenz/Artigenz-Coder-DS-6.7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Artigenz/Artigenz-Coder-DS-6.7B")
model = AutoModelForMultimodalLM.from_pretrained("Artigenz/Artigenz-Coder-DS-6.7B")
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 Settings

vLLM

How to use Artigenz/Artigenz-Coder-DS-6.7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Artigenz/Artigenz-Coder-DS-6.7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Artigenz/Artigenz-Coder-DS-6.7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Artigenz/Artigenz-Coder-DS-6.7B

SGLang

How to use Artigenz/Artigenz-Coder-DS-6.7B 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 "Artigenz/Artigenz-Coder-DS-6.7B" \
    --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": "Artigenz/Artigenz-Coder-DS-6.7B",
		"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 "Artigenz/Artigenz-Coder-DS-6.7B" \
        --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": "Artigenz/Artigenz-Coder-DS-6.7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Artigenz/Artigenz-Coder-DS-6.7B with Docker Model Runner:
```
docker model run hf.co/Artigenz/Artigenz-Coder-DS-6.7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Artigenz-Coder-DS-6.7B

Artigenz team intends to create family of code generation models that can run very fast on local computers.

Artigenz-Coder-DS-6.7B is the first in this family with 6.7B parameters and 13GB memory footprint 🌟

HomePage

About the model

Artigenz-Coder-DS-6.7B was finetuned on DeepSeek-Coder-6.7B-Base. The dataset and scripts will be open-sourced soon.

We have open sourced our model weights on 🤗 HF, checkout here!

Team

Nikita Agarwal

AI Researcher

ex Data Scientist at Microsoft

IIIT - Hyderabad, India

Vivek Verma

Post Doctoral Associate

Florida International Univesity

202 Citations

Nalin Abrol

ex Software Engineer - Plivo (YC S21)

Published in OHBM 2019↗

IIIT - Hyderabad, India

Special Thanks ❤️

Manish Shrivastava

Assistant Professor

Natural Language Processing

IIIT - Hyderabad, India

Manas Kumar Verma

CEO

Algouniversity YC(S21)

IIIT - Hyderabad, India

Nikhil Tadigoppula

AI Researcher

Bronze medalist

International Olympiad

in Informatics 2013

IIIT - Hyderabad, India

What's Next ❓

The dataset and finetuing scripts used to train Artigenz-Coder-DS-6.7B will be released soon for the open-source-community to use freely. 🛠️.

1B & 3B models from Artigenz family are on the roadmap next with long term goal to enable ⚡ fast local inference for code generation.

Special Thanks to the Open Source Community ❤️

We extend our deepest gratitude to the open source community, especially the Bigcode Project, Magicoder, Hugging Face, DeepSeek, Wizard Coder, Code Llama that enabled research community to build powerfull LLMs.

We need many more people to close the gap between proprietry and open source models and we are commited to contribute our bits to the goal.