Instructions to use swap-uniba/LLaMAntino-2-chat-13b-hf-ITA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use swap-uniba/LLaMAntino-2-chat-13b-hf-ITA with Transformers:

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

pipe = pipeline("text-generation", model="swap-uniba/LLaMAntino-2-chat-13b-hf-ITA")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("swap-uniba/LLaMAntino-2-chat-13b-hf-ITA")
model = AutoModelForCausalLM.from_pretrained("swap-uniba/LLaMAntino-2-chat-13b-hf-ITA", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use swap-uniba/LLaMAntino-2-chat-13b-hf-ITA with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/swap-uniba/LLaMAntino-2-chat-13b-hf-ITA

SGLang

How to use swap-uniba/LLaMAntino-2-chat-13b-hf-ITA 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 "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA" \
    --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": "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA",
		"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 "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA" \
        --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": "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use swap-uniba/LLaMAntino-2-chat-13b-hf-ITA with Docker Model Runner:
```
docker model run hf.co/swap-uniba/LLaMAntino-2-chat-13b-hf-ITA
```

Model Card for LLaMAntino-2-chat-13b-ITA

Model description

LLaMAntino-2-chat-13b is a Large Language Model (LLM) that is an italian-adapted LLaMA 2 chat. This model aims to provide Italian NLP researchers with a base model for italian dialogue use cases.

The model was trained using QLora and using as training data clean_mc4_it medium. If you are interested in more details regarding the training procedure, you can find the code we used at the following link:

Repository: https://github.com/swapUniba/LLaMAntino

NOTICE: the code has not been released yet, we apologize for the delay, it will be available asap!

Developed by: Pierpaolo Basile, Elio Musacchio, Marco Polignano, Lucia Siciliani, Giuseppe Fiameni, Giovanni Semeraro
Funded by: PNRR project FAIR - Future AI Research
Compute infrastructure: Leonardo supercomputer
Model type: LLaMA 2 chat
Language(s) (NLP): Italian
License: Llama 2 Community License
Finetuned from model: NousResearch/Llama-2-13b-chat-hf

How to Get Started with the Model

Below you can find an example of model usage:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "swap-uniba/LLaMAntino-2-chat-13b-hf-ITA"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

prompt = "Scrivi qui un possibile prompt"

input_ids = tokenizer(prompt, return_tensors="pt").input_ids
outputs = model.generate(input_ids=input_ids)

print(tokenizer.batch_decode(outputs.detach().cpu().numpy()[:, input_ids.shape[1]:], skip_special_tokens=True)[0])

If you are facing issues when loading the model, you can try to load it quantized:

model = AutoModelForCausalLM.from_pretrained(model_id, load_in_8bit=True)

Note: The model loading strategy above requires the bitsandbytes and accelerate libraries

Citation

If you use this model in your research, please cite the following:

@misc{basile2023llamantino,
      title={LLaMAntino: LLaMA 2 Models for Effective Text Generation in Italian Language}, 
      author={Pierpaolo Basile and Elio Musacchio and Marco Polignano and Lucia Siciliani and Giuseppe Fiameni and Giovanni Semeraro},
      year={2023},
      eprint={2312.09993},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Downloads last month: 12

Safetensors

Model size

13B params

Tensor type

F32

F16

Collection including swap-uniba/LLaMAntino-2-chat-13b-hf-ITA

LLaMAntino Models

Collection

Collection of all the published LLaMAntino models • 10 items • Updated May 14, 2024 • 5

Paper for swap-uniba/LLaMAntino-2-chat-13b-hf-ITA

LLaMAntino: LLaMA 2 Models for Effective Text Generation in Italian Language

Paper • 2312.09993 • Published Dec 15, 2023 • 4