Instructions to use MaralGPT/Maral-7B-alpha-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MaralGPT/Maral-7B-alpha-1 with Transformers:

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

pipe = pipeline("text-generation", model="MaralGPT/Maral-7B-alpha-1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MaralGPT/Maral-7B-alpha-1")
model = AutoModelForCausalLM.from_pretrained("MaralGPT/Maral-7B-alpha-1", device_map="auto")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MaralGPT/Maral-7B-alpha-1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MaralGPT/Maral-7B-alpha-1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MaralGPT/Maral-7B-alpha-1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MaralGPT/Maral-7B-alpha-1

SGLang

How to use MaralGPT/Maral-7B-alpha-1 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 "MaralGPT/Maral-7B-alpha-1" \
    --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": "MaralGPT/Maral-7B-alpha-1",
		"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 "MaralGPT/Maral-7B-alpha-1" \
        --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": "MaralGPT/Maral-7B-alpha-1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MaralGPT/Maral-7B-alpha-1 with Docker Model Runner:
```
docker model run hf.co/MaralGPT/Maral-7B-alpha-1
```

thanks

by LittenBuzz - opened Dec 25, 2023

Discussion

LittenBuzz

Dec 25, 2023

Hi
I wanted to thank you for your efforts on providing this model. It would be more accessible if you could also provide GPTQ quantization 🙏🏻

Muhammadreza

MaralGPT org Dec 26, 2023

Well, we're working on it. It'll be released soon.

Arena

Dec 27, 2023

Mohammadreza, thank you for making this model available. I'm keenly interested in contributing to Persian Large Language Models (LLMs) as a native speaker. Would you consider any opportunity to collaborate?

Muhammadreza

MaralGPT org Dec 27, 2023

@Arena , why not. I guess we need to create a community, on a platform such as Telegram.

Arena

Dec 27, 2023

Certainly, it's a nice first step.

Muhammadreza

MaralGPT org Dec 28, 2023

@Arena , I will make a Telegram group for further developments. I guess we can talk there more. Please feel free to message me on haghiri75@gmail.com for accessing the community.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment