Instructions to use intelia-lab-uah/mT5-base_AE_SS with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use intelia-lab-uah/mT5-base_AE_SS with Transformers:

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

pipe = pipeline("text-generation", model="intelia-lab-uah/mT5-base_AE_SS")

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("intelia-lab-uah/mT5-base_AE_SS")
model = AutoModelForSeq2SeqLM.from_pretrained("intelia-lab-uah/mT5-base_AE_SS")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use intelia-lab-uah/mT5-base_AE_SS with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "intelia-lab-uah/mT5-base_AE_SS"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "intelia-lab-uah/mT5-base_AE_SS",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/intelia-lab-uah/mT5-base_AE_SS

SGLang

How to use intelia-lab-uah/mT5-base_AE_SS 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 "intelia-lab-uah/mT5-base_AE_SS" \
    --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": "intelia-lab-uah/mT5-base_AE_SS",
		"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 "intelia-lab-uah/mT5-base_AE_SS" \
        --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": "intelia-lab-uah/mT5-base_AE_SS",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use intelia-lab-uah/mT5-base_AE_SS with Docker Model Runner:
```
docker model run hf.co/intelia-lab-uah/mT5-base_AE_SS
```

YAML Metadata Warning:The pipeline tag "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

Modifications and Derivative Work Notice

This model is based on google/mt5-base, licensed under the Apache License 2.0.

This repository contains a modified and fine-tuned version of the original model.

Modifications include:

Additional training on the SQuAD+SQAC dataset to finetune the model on the Answer Extraction task
Hyperparameter adjustments

A detailed description of the modifications, training procedure, and experimental setup can be found in the associated paper: Evaluating the performance of multilingual models in answer extraction and question generation.

All modifications were made by INTELIA.

Citation

If you use this model in your research, please cite the following paper: Evaluating the performance of multilingual models in answer extraction and question generation.

@article{moreno-cediel_evaluating_2024,
  title = {Evaluating the performance of multilingual models in answer extraction and question generation},
  volume = {14},
  copyright = {2024 The Author(s)},
  issn = {2045-2322},
  url = {https://www.nature.com/articles/s41598-024-66472-5},
  doi = {10.1038/s41598-024-66472-5},
  language = {en},
  number = {1},
  urldate = {2025-01-10},
  journal = {Scientific Reports},
  author = {Moreno-Cediel, Antonio and del-Hoyo-Gabaldon, Jesus-Angel and Garcia-Lopez, Eva and Garcia-Cabot, Antonio and de-Fitero-Dominguez, David},
  month = jul,
  year = {2024},
  note = {Publisher: Nature Publishing Group},
  keywords = {Computer science, Software},
  pages = {15477},
}