Instructions to use gplsi/Aitana-2B-SI-Instruct-Aligned with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use gplsi/Aitana-2B-SI-Instruct-Aligned with Transformers:

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

pipe = pipeline("text-generation", model="gplsi/Aitana-2B-SI-Instruct-Aligned")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("gplsi/Aitana-2B-SI-Instruct-Aligned")
model = AutoModelForCausalLM.from_pretrained("gplsi/Aitana-2B-SI-Instruct-Aligned")
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

vLLM

How to use gplsi/Aitana-2B-SI-Instruct-Aligned with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gplsi/Aitana-2B-SI-Instruct-Aligned"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "gplsi/Aitana-2B-SI-Instruct-Aligned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/gplsi/Aitana-2B-SI-Instruct-Aligned

SGLang

How to use gplsi/Aitana-2B-SI-Instruct-Aligned 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 "gplsi/Aitana-2B-SI-Instruct-Aligned" \
    --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": "gplsi/Aitana-2B-SI-Instruct-Aligned",
		"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 "gplsi/Aitana-2B-SI-Instruct-Aligned" \
        --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": "gplsi/Aitana-2B-SI-Instruct-Aligned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use gplsi/Aitana-2B-SI-Instruct-Aligned with Docker Model Runner:
```
docker model run hf.co/gplsi/Aitana-2B-SI-Instruct-Aligned
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Aitana-2B-SI-Instruct-Aligned

Aitana-2B-SI-Instruct-Aligned is a DPO-aligned instruction-tuned generative language model from the Aitana family, developed by the GPLSI (Language and Information Systems Group) at the University of Alicante. Built on gplsi/Aitana-2B-SI-Instruct, this model has been further aligned using Direct Preference Optimization (DPO) to improve response quality and alignment with human preferences across Valencian, Spanish, and English.

Model Description
Alignment Details
Training Data
Intended Uses
How to Use
Evaluation
Additional Information

Model Description

Property	Value
Base Model	gplsi/Aitana-2B-SI-Instruct
Architecture	Transformer decoder-only
Parameters	~2.25B
Languages	Valencian, Spanish, English
License	Apache 2.0

Aitana-2B-SI-Instruct-Aligned extends the Aitana-2B-SI-Instruct instruction-tuned model with Direct Preference Optimization (DPO) alignment. This additional training stage improves the model's ability to generate helpful, high-quality responses that better align with human preferences while maintaining its strong multilingual capabilities.

Alignment Details

The model was aligned using Direct Preference Optimization (DPO) with the following configuration:

Hyperparameter	Value
Method	DPO (Direct Preference Optimization)
Learning rate	5e-6
Epochs	1
Beta	0.1
LR Scheduler	Linear
Total Samples	146,180
English Samples	80,308
Spanish Samples	30,072
Valencian Samples	35,800
Languages	Spanish, Valencian, English

The DPO alignment was performed using curated preference pairs that teach the model to prefer more helpful, accurate, and well-structured responses.

Training Data

The base instruction model was trained on the ALIA Instruction/v12 dataset. This DPO-aligned variant was further aligned using the Alignment/v8 dataset, composed of the following preference data:

Dataset ID	Name	Languages	Source
al1	HelpSteer3	EN, ES	nvidia/HelpSteer3
al2	OpenAssistant1 (OASST1)	EN, ES, RU (+32 more)	OpenAssistant/oasst1
al3	OpenAssistant2 (OASST2)	EN, ES, RU (+32 more)	OpenAssistant/oasst2
al4	OpenOrca	EN	Open-Orca/OpenOrca
al5	OASST2 Valenciano	VA	—

The alignment data focused on English, Spanish, and Valencian preference pairs, with the distribution: 80,308 English, 30,072 Spanish, and 35,800 Valencian samples.

Intended Uses

This model can be used for:

Instruction following in Valencian, Spanish, and English with improved alignment to human preferences
Chat and conversational applications requiring high-quality multilingual responses
Text generation with task-specific prompting and improved output quality
Domain-specific applications in administrative, legal, or tourism contexts

Note: As an aligned instruction-tuned model, it is designed to follow user prompts and generate helpful, safe responses. It is not intended for use as a factual knowledge base. The DPO alignment improves response quality and preference alignment.

How to Use

Transformers

import torch
from transformers import pipeline, AutoTokenizer

model_id = "gplsi/Aitana-2B-SI-Instruct-Aligned"
tokenizer = AutoTokenizer.from_pretrained(model_id)
generator = pipeline(
    "text-generation",
    model=model_id,
    tokenizer=tokenizer,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
# Valencian example
text = "Explica què són les Corts Valencianes i quina funció tenen."
result = generator(text, do_sample=True, top_k=10, max_new_tokens=100)
print(result[0]['generated_text'])
# Spanish example
text = "Describe las principales funciones del gobierno autonómico valenciano."
result = generator(text, do_sample=True, top_k=10, max_new_tokens=100)
print(result[0]['generated_text'])
# English example
text = "Explain the role of tourism in the Valencian Community economy."
result = generator(text, do_sample=True, top_k=10, max_new_tokens=100)
print(result[0]['generated_text'])

Evaluation

In the following tables, we present the results obtained with different benchmarks from lm-evaluation-harness in comparison with Salamandra-2B-Instruct, and Aitana-2B-S-Instruct-Aligned. The results reflect the DPO-aligned instruction-tuned performance.

Valencian

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
XNLI	va	Natural Language Inference	acc	0.520	0.514	0.485

Generation Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
Cocoteros	va	Reading Comprehension	bleu	2.796	3.612	4.223
Phrases ca-va	va-ca	Translation - Adaptation	bleu	58.425	74.538	68.305
Phrases va-ca	va-ca	Translation - Adaptation	bleu	70.660	71.691	69.551
Phrases va-es	va-es	Translation	bleu	65.427	72.097	70.061
Phrases es-va	es-va	Translation	bleu	45.688	56.012	54.053
Truthfulqa_va	va	Truthfulness	bleu_acc	0.409	0.394	0.383

Catalan

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
Belebele Cat_latn	ca	Reading Comprehension	acc	0.287	0.248	0.319
COPA	ca	Commonsense Reasoning	acc	0.708	0.726	0.694
XStoryCloze	ca	Commonsense Reasoning	acc	0.616	0.629	0.623
OpenBookQA	ca	Question Answering	acc	0.296	0.296	0.326
PAWS	ca	Paraphrasing	acc	0.602	0.598	0.531
PiQA	ca	Question Answering	acc	0.638	0.655	0.629
ARC Easy	ca	Question Answering	acc	0.516	0.524	0.526
ARC Challenge	ca	Question Answering	acc	0.298	0.314	0.310
XNLI	ca	Natural Language Inference	acc	0.513	0.515	0.497
Teca	ca	Natural Language Inference	acc	0.486	0.500	0.468
WNLI	ca	Natural Language Inference	acc	0.563	0.437	0.436
Catcola	ca	Linguistic Acceptability	acc	0.492	0.713	0.680
Catcola	ca	Linguistic Acceptability	mcc	0.097	-0.040	0.013
Catalanqa	ca	Question Answering	F1	0.516	0.384	0.396
Mgsm direct	ca	Math	exact match	0.000	0.012	0.004
Catalanqa	ca	Question Answering	exact match	0.182	0.011	0.031
Xquad	ca	Question Answering	exact match	0.103	0.014	0.037
Xquad	ca	Question Answering	F1	0.394	0.287	0.317

Generation Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
Cabreu abstractive	ca	Summarization	bleu	7.610	7.703	8.837
Cabreu extractive	ca	Summarization	bleu	38.002	19.876	28.16803
Cabreu extreme	ca	Summarization	bleu	2.733	3.245	3.386

Spanish

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
Belebele	es	Reading Comprehension	acc	0.268	0.244	0.285
PAWS	es	Paraphrasing	acc	0.566	0.618	0.546
XNLI	es	Natural Language Inference	acc	0.463	0.439	0.443
WNLI	es	Natural Language Inference	acc	0.479	0.535	0.535
XStoryCloze	es	Commonsense Reasoning	acc	0.617	0.628	0.632
Escola	es	Linguistic Acceptability	acc	0.293	0.708	0.654
Escola	es	Linguistic Acceptability	mcc	0.020	0.000	0.046
OpenbookQA	es	Question Answering	acc	0.286	0.338	0.332
MGSM Direct	es	Math	exact match	0.020	0.024	0.1
XQUAD	es	Question Answering	exact match	0.066	0.026	0.019
XQUAD	es	Question Answering	F1	0.355	0.293	0.293

Generation Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
Cocoteros	es	Reading Comprehension	bleu	3.308	3.141	3.670
XLSum	es	Summarization	bleu	1.695	1.737	1.971

English

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
Arc Challenge	en	Question Answering	acc	0.354	0.363	0.372
Arc Easy	en	Question Answering	acc	0.681	0.709	0.682
Belebele	en	Reading Comprehension	acc	0.260	0.293	0.349
PAWS	en	Paraphrasing	acc	0.597	0.594	0.555
XNLI	en	Natural Language Inference	acc	0.512	0.553	0.480
XStoryCloze	en	Commonsense Reasoning	acc	0.662	0.680	0.693
OpenBookQA	en	Question Answering	acc	0.298	0.338	0.316
PiQA	en	Question Answering	acc	0.715	0.717	0.704
Social iqa	en	Question Answering	acc	0.453	0.451	0.468
WNLI	en	Natural Language Inference	acc	0.535	0.465	0.451
MGSM Direct	en	Math	exact match	0.008	0.052	0.116
TriviaQA	en	Question Answering	exact match	0.076	0.147	0.156

Judge Evaluation

The model was also evaluated using an LLM-as-judge approach across different task categories. The scores below represent the average rating (1-5 scale, 5 being best) and standard deviation for each task category, comparing Aitana-2B-SI-Instruct-Aligned against Salamandra-2B-Instruct and Aitana-2B-S-Instruct-Aligned.

Task Category	Salamandra-2B-Instruct	Aitana-2B-S-Instruct-Aligned (v0.1)	Aitana-2B-SI-Instruct-Aligned (v0.1)
CommonSense reasoning	2.277 / 1.151	2.737 / 1.140	2.969 / 1.086
Maths	1.060 / 0.124	1.123 / 0.249	1.191 / 0.349
Paraphrasing	3.518 / 1.308	3.460 / 1.088	3.472 / 0.959
Reading comprehension	2.966 / 1.111	2.894 / 1.311	3.112 / 1.146
Summarization	2.217 / 1.068	2.261 / 0.820	2.591 / 1.115
Translation	3.557 / 0.760	3.418 / 0.999	3.390 / 0.730
Overall Avg	2.599 / 0.920	2.649 / 0.935	2.787 / 0.897

The DPO-aligned model shows a notable improvement in overall average score (2.787) compared to Aitana-2B-S-Instruct-Aligned (v0.1) (2.649) and Salamandra-2B-Instruct (2.599) with particular gains in CommonSense reasoning, reading comprehension and summarization. The aligned model also shows tighter standard deviations in several categories, indicating more consistent quality responses.

Additional Information

Author

The model has been developed by the Language and Information Systems Group (GPLSI) and the Centro de Inteligencia Digital (CENID), both part of the University of Alicante (UA), as part of their ongoing research in Natural Language Processing (NLP).

Funding

This work is funded by the Ministerio para la Transformación Digital y de la Función Pública, co-financed by the EU – NextGenerationEU, within the framework of the project Desarrollo de Modelos ALIA. This work has also been partially supported by Project HEART-NLP (PID2024-156263OB-C22).

Acknowledgments

We would like to express our gratitude to all individuals and institutions that have contributed to the development of this work. Special thanks to:

We also acknowledge the financial, technical, and scientific support of the Ministerio para la Transformación Digital y de la Función Pública - Funded by EU – NextGenerationEU within the framework of the project Desarrollo de Modelos ALIA, whose contribution has been essential to the completion of this research.

License

Apache License, Version 2.0

Disclaimer

This model is intended for general purposes and is available under a permissive Apache License 2.0. Be aware that the model may have biases and/or undesirable outputs. Users deploying systems based on this model are responsible for mitigating risks and complying with applicable AI regulations.

Reference

@misc{gplsi-aitana-2B-SI-Instruct-Aligned,
  author       = {Galiano, Santiago and Sepúlveda-Torres, Robiert and Martínez-Murillo, Iván and Grande, Eduardo and Consuegra-Ayala, Juan Pablo and Miró Maestre, María and Canal-Esteve, Miquel and Bonora, Mar and Gutierrez, Yoan and Abreu Salas, José Ignacio and Lloret, Elena and Montoyo, Andrés and Muñoz-Guillena, Rafael and Palomar, Manuel},
  title        = {Aitana 2B SI Instruct Aligned: DPO-aligned instruction-tuned model for Valencian, Spanish and English},
  year         = {2026},
  institution  = {Language and Information Systems Group (GPLSI) and Centro de Inteligencia Digital (CENID), University of Alicante (UA)},
  howpublished = {\url{https://huggingface.co/gplsi/Aitana-2B-SI-Instruct-Aligned}},
  note         = {Accessed: 2026-05-11}
}

Downloads last month: 83

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for gplsi/Aitana-2B-SI-Instruct-Aligned

Base model

gplsi/Aitana-2B-SI-Instruct

Finetuned

(1)

this model

Quantizations

1 model

gplsi
/

Aitana-2B-SI-Instruct-Aligned

Aitana-2B-SI-Instruct-Aligned

Table of Contents

Model Description

Alignment Details

Training Data

Intended Uses

How to Use

Transformers

Evaluation

Valencian

Classification Benchmarks

Generation Benchmarks

Catalan

Classification Benchmarks

Generation Benchmarks

Spanish

Classification Benchmarks

Generation Benchmarks

English

Classification Benchmarks

Judge Evaluation

Additional Information

Author

Funding

Acknowledgments

License

Disclaimer

Reference

Model tree for gplsi/Aitana-2B-SI-Instruct-Aligned

Datasets used to train gplsi/Aitana-2B-SI-Instruct-Aligned