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

Libraries

How to use gplsi/Aitana-2B-S-tourism-Instruct with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("gplsi/Aitana-2B-S-tourism-Instruct")
model = AutoModelForMultimodalLM.from_pretrained("gplsi/Aitana-2B-S-tourism-Instruct")
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 gplsi/Aitana-2B-S-tourism-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gplsi/Aitana-2B-S-tourism-Instruct"
# 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-S-tourism-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/gplsi/Aitana-2B-S-tourism-Instruct

SGLang

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

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

Aitana-2B-S-tourism-Instruct

Aitana-2B-S-tourism-Instruct is an 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-S-tourism-base, this model has been fine-tuned to follow instructions across Valencian, Spanish, and English, with specialized capabilities for tourism domain applications.

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

Model Description

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

Aitana-2B-S-tourism-Instruct extends the Aitana-2B-S-tourism-base domain-specific foundation model with instruction fine-tuning. This combination makes it particularly well-suited for tourism-related tasks requiring instruction following in Valencian, Spanish, and English.

Training Data

This model was instruction fine-tuned using the following data:

Dataset ID	Name	Languages	Source
ins1	InstruCAT	CA	projecte-aina/InstruCAT
ins2	NLUCat	CA	projecte-aina/NLUCat
ins3	Escagleu 64K	CA	projecte-aina/escagleu-64k
ins4	OpenAssistant2 (OASST2)	CA, EN, ES, VA	OpenAssistant/oasst2
ins5	OpenAssistant1 (OASST1)	CA, VA	projecte-aina/oasst1_ca
ins6	M-Personas	CA, EN, ES, VA	BSC-LT/m-personas
ins7	RAG Multilingual	CA, EN, ES	projecte-aina/RAG_Multilingual
ins8	FLORES	CA, EN, ES	facebook/flores
ins9	Aya Dataset	EN, ES, VA	CohereLabs/aya_dataset
ins10	TowerBlocks	EN, ES	Unbabel/TowerBlocks-v0.1
ins11	Mentor / Mentores	CA, ES, VA	projecte-aina/MentorES / projecte-aina/MentorCA
ins12	Dolly / Dolly 3K	CA, EN, VA	databricks/databricks-dolly-15k / projecte-aina/dolly3k_ca
ins13	Alpaca	EN, VA	yahma/alpaca-cleaned
ins14	GSM8K	EN, VA	openai/gsm8k
ins15	OpenOrca	EN	Open-Orca/OpenOrca
ins16	No Robots	EN	HuggingFaceH4/no_robots
ins17	TableGPT	EN	LipengCS/Table-GPT
ins18	CoQCA / CoQCat	CA, VA	projecte-aina/CoQCat
ins19	SciFact	EN, VA	allenai/scifact
ins20	LingComp QA	ES, VA	somosnlp/LingComp_QA
ins21	Instruct Legal Refugiados	ES, VA	somosnlp/instruct-legal-refugiados-es
ins22	Gastronomia Hispana	ES, VA	somosnlp-hackathon-2025/gastronomia-hispana-dpo
ins23	TurismInstructionsGPLSI	VA	—
ins24	Amic-Paralelo	VA	—
ins25	BOUA	VA	gplsi/boua_parallel
ins26	DOGV Parallel	VA	—
ins27	UJI VA-EN Parallel	VA	—
ins28	UJI VA-ES Parallel	VA	—

Intended Uses

This model can be used for:

Tourism text generation in Valencian, Spanish, and English
Travel content creation and visitor assistance
Instruction following with tourism domain expertise
Fine-tuning for specific tourism downstream tasks

Note: This model combines tourism domain specialization with instruction-following capabilities. For general-purpose instruction following, consider other models in the Aitana family.

How to Use

Transformers

import torch
from transformers import pipeline, AutoTokenizer

model_id = "gplsi/Aitana-2B-S-tourism-Instruct"
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 = "Recomana'm les millors platges de la Costa Blanca per a unes vacances familiars."
result = generator(text, do_sample=True, top_k=10, max_new_tokens=100)
print(result[0]['generated_text'])
# Spanish example
text = "Describe los principales atractivos turísticos de la Comunidad Valenciana."
result = generator(text, do_sample=True, top_k=10, max_new_tokens=100)
print(result[0]['generated_text'])
# English example
text = "What are the best cultural sites to visit in Valencia?"
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.

Normalized score per language

Language	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Spanish	0.079	0.086
Catalan	0.202	0.177
English	0.178	0.164
Valencian	0.507	0.483
Average	0.242	0.228

Valencian

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
XNLI	va	Natural Language Inference	acc	0.520	0.483

Generation Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Cocoteros	va	Reading Comprehension	bleu	2.796	3.414
Phrases ca-va	va-ca	Translation - Adaptation	bleu	58.425	70.188
Phrases va-ca	va-ca	Translation - Adaptation	bleu	70.660	66.078
Phrases va-es	va-es	Translation	bleu	65.427	41.781
Phrases es-va	es-va	Translation	bleu	45.688	46.205
Truthfulqa_va	va	Truthfulness	bleu_acc	0.409	0.377

Catalan

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Belebele Cat_latn	ca	Reading Comprehension	acc	0.287	0.254
COPA	ca	Commonsense Reasoning	acc	0.708	0.710
XStoryCloze	ca	Commonsense Reasoning	acc	0.616	0.621
OpenBookQA	ca	Question Answering	acc	0.296	0.276
PAWS	ca	Paraphrasing	acc	0.602	0.600
PiQA	ca	Question Answering	acc	0.638	0.639
SiQA	ca	Question Answering	acc	0.422	0.428
ARC Easy	ca	Question Answering	acc	0.516	0.495
ARC Challenge	ca	Question Answering	acc	0.298	0.311
XNLI	ca	Natural Language Inference	acc	0.513	0.494
Teca	ca	Natural Language Inference	acc	0.486	0.487
WNLI	ca	Natural Language Inference	acc	0.563	0.437
Catcola	ca	Linguistic Acceptability	acc	0.492	0.663
Catcola	ca	Linguistic Acceptability	mcc	0.097	0.011
Catalanqa	ca	Question Answering	F1	0.516	0.372
Mgsm direct	ca	Math	exact match	0.000	0.000
Catalanqa	ca	Question Answering	exact match	0.182	0.029
Xquad	ca	Question Answering	exact match	0.103	0.032
Xquad	ca	Question Answering	F1	0.394	0.290

Generation Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Cabreu abstractive	ca	Summarization	bleu	7.610	8.250
Cabreu extractive	ca	Summarization	bleu	38.002	31.959
Cabreu extreme	ca	Summarization	bleu	2.733	3.168

Spanish

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Belebele	es	Reading Comprehension	acc	0.268	0.240
PAWS	es	Paraphrasing	acc	0.566	0.609
XNLI	es	Natural Language Inference	acc	0.463	0.394
WNLI	es	Natural Language Inference	acc	0.479	0.437
XStoryCloze	es	Commonsense Reasoning	acc	0.617	0.614
Escola	es	Linguistic Acceptability	acc	0.293	0.544
Escola	es	Linguistic Acceptability	mcc	0.020	0.029
OpenbookQA	es	Question Answering	acc	0.286	0.296
MGSM Direct	es	Math	exact match	0.020	0.068
XQUAD	es	Question Answering	exact match	0.066	0.018
XQUAD	es	Question Answering	F1	0.355	0.282

Generation Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Cocoteros	es	Reading Comprehension	bleu	3.308	2.545
XLSum	es	Summarization	bleu	1.695	1.472

English

Classification Benchmarks

Dataset	Lang.	Task	Metric	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
Arc Challenge	en	Question Answering	acc	0.354	0.336
Arc Easy	en	Question Answering	acc	0.681	0.668
Belebele	en	Reading Comprehension	acc	0.260	0.243
PAWS	en	Paraphrasing	acc	0.597	0.623
XNLI	en	Natural Language Inference	acc	0.512	0.551
XStoryCloze	en	Commonsense Reasoning	acc	0.662	0.666
OpenBookQA	en	Question Answering	acc	0.298	0.296
PiQA	en	Question Answering	acc	0.715	0.726
Social iqa	en	Question Answering	acc	0.453	0.420
WNLI	en	Natural Language Inference	acc	0.535	0.423
MGSM Direct	en	Math	exact match	0.008	0.056
TriviaQA	en	Question Answering	exact match	0.076	0.051

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 against Salamandra-2B-Instruct.

Task Category	Salamandra-2B-Instruct	Aitana-2B-S-tourism-Instruct
CommonSense reasoning	2.277 / 1.151	1.962 / 1.010
Maths	1.060 / 0.124	1.079 / 0.146
Paraphrasing	3.518 / 1.308	3.547 / 1.199
Reading comprehension	2.966 / 1.111	2.649 / 1.303
Summarization	2.217 / 1.068	1.961 / 0.875
Translation	3.557 / 0.760	3.494 / 1.052
Overall Avg	2.599 / 0.920	2.448 / 0.931

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-S-tourism-Instruct,
  author       = {Martínez-Murillo, Iván and Sepúlveda-Torres, Robiert and Grande, Eduardo and Galiano, Santiago and Estevanell-Valladares, Ernesto L. 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 Tourism Instruct: Instruction-tuned model for tourism applications in 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-S-tourism-Instruct}},
  note         = {Accessed: 2026-05-21}
}

Downloads last month: 113

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for gplsi/Aitana-2B-S-tourism-Instruct

Unable to build the model tree, the base model loops to the model itself. Learn more.

gplsi
/

Aitana-2B-S-tourism-Instruct

Aitana-2B-S-tourism-Instruct

Table of Contents

Model Description

Training Data

Intended Uses

How to Use

Transformers

Evaluation

Normalized score per language

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-S-tourism-Instruct

Datasets used to train gplsi/Aitana-2B-S-tourism-Instruct