Model Card for RigoChat-7b-v2

RigoChat-7b-v2 is a Qwen-2.5-based model specifically designed to provide accurate responses from Spanish queries. Specifically, is based on the Qwen/Qwen2.5-7B-Instruct model and has been fine-tuned with Direct Preference Optimization (DPO) for improved performance in Spanish language.

This model is licensed for non-commercial use. If you want to use it commercially, please contact us or use it through the service we offer from the AWS Marketplace.

Model Details

Model Description

This model is the second version of RigoChat, a family of Large Language Models (LLMs) designed to solve typical NLP tasks with Spanish instructions such as: Tool Use, Summarization, Math, Code, Abstractive-QA, etc. Like Qwen/Qwen2.5-7B-Instruct, this model has no specific use case and can be applied to a wide range of tasks. Indeed, it offers a slight improvement for generalist tasks in Spanish, particularly in RAG (Retriever Augmented Generation) systems with Spanish databases, as its training focused on resolving questions about contexts to prevent hallucinations and ensure safe responses.

Key benefits of this model include:

Improved performance on generalist tasks in Spanish.
Enhanced safety and reduced hallucinations in RAG systems with Spanish texts.
Possibility of using it in different hardware requirements, especially those with reduced computational capacity. For more information on how to use RigoChat-7b-v2 on reduced hardware, see IIC/RigoChat-7b-v2-GGUF.

Remarkably, this model was trained on a single A100 GPU with limited computational resources, yet achieved its current state in a relatively short time (8.5 hours). This feat was made possible by leveraging a high-quality dataset and employing advanced techniques such as LoRA to optimize memory usage. Further details on the training process can be found below.

Developed by: Instituto de Ingeniería del Conocimiento (IIC).
Model type: Generative Fine-tuned Transformer.
Language(s) (NLP): Spanish (BCP-47 es).
License: RIGOCHAT NON-COMMERCIAL.
Arquitecture: We use Qwen's architecture without modifications.
Finetuned from model: Qwen/Qwen2.5-7B-Instruct.

Model Sources

Paper: Coming soon.

Uses

Direct Use

You can use and deploy RigoChat-v2 for commercial purposes through a model package from AWS Marketplace. You can check the instructions inside the following notebook.

Out-of-Scope Use

This language model has been adapted for general natural language processing tasks in Spanish and specific use cases such as RAG. However, there are several cases where the model should not be used due to its technical and ethical limitations:

Illegal Activities: The model should not be used to generate content related to illegal activities, such as creating malicious software, fraud, incitement to crime, or any illegal material.
Harmful or Dangerous Content: It should not be used to generate hate speech, violence, harassment, or any content that promotes discrimination, violence, or abuse.

Bias, Risks, and Limitations

Although this model has been trained to understand and generate text in Spanish, there are several risks, biases, and limitations that users should be aware of:

Biases: The model may reflect biases present in the training data. These biases could be related to gender, race, social class, sexual orientation, among others, and may generate responses that perpetuate stereotypes or discrimination.
Accuracy and Reliability: While the model generates coherent and useful text in many contexts, it may not always be 100% accurate or reliable, especially in technical, scientific, or legal matters where high certainty is required.
Limited or Outdated Knowledge: The model is not trained with information beyond its training cutoff date. Therefore, it may not reflect recent events, research, or advancements.

Recommendations

We recommend using this model as a general chatbot or within applications designed for specific tasks, such as SQL queries, RAG systems, or as an autonomous agent to facilitate the use of tools.

How to Get Started with the Model

As Qwen/Qwen2.5-7B-Instruct or any model that uses the chatml template.

To load the model and tokenizer

from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
)
import torch

model_name = "IIC/RigoChat-7b-v2"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="cuda",
    trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(
    model_name,
    trust_remote_code=True,
)

Sample generation

messages = [
    {"role": "user", "content": "¿Cómo puedo transformar un diccionario de listas en una lista de diccionarios, y viceversa, en Python sin utilizar bucles for?"}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=1024,
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

For a better experience, we recommend to use the following generation parameters.

Tool Use

def get_current_weather(location: str, date: str) -> float:
    """
    Obtener datos del tiempo de una localización.
    
    Args:
        location: La locaización, con el siguiente formato: "Ciudad, País."
        date: La fecha, en el formato AAAA-MM-DD.
    Returns:
        El tiempo en dicha localización.
    """
    return {"temperatura": 22, "cielo": "nublado", "probabilidad de lluvias": "60%"}

messages = [
  {"role": "user", "content": "Este fin de semana quiero visitar Madrid, y no se qué ropa llevarme. ¿Podrías decirme qué tal va a hacer? Es el puente del 6 de diciembre de 2024."}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    tools=[get_current_weather],
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=1024
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Check the tool use documentation from HuggingFace for more information.

Training Details

Training Data

A combination of both public and private datasets designed in the IIC. The dataset consists of 21975 conversations in Spanish, with the format chatml and has the same structure as the Anthropic/hh-rlhf dataset. Each conversation has two variants: chosen and rejected, and only differs the last answer of the assistant. The last answer in the chosen variant is considered a better answer than the one in the rejected variant. Different techniques have been used to generate the dataset, which we explain in depth in the research (coming soon).

Training Procedure

We use the Transformer Reinforcement Learning (TRL) library. Specifically, we have applied the script they have published as an example for using DPO to the dataset we have generated.

Training Hyperparameters

Details

LORA_CONFIG = {
    "r": 64,
    "lora_alpha": 16,
    "lora_dropout": 0.1,
    "bias": "none",
    "task_type": "CAUSAL_LM",
    "target_modules": [
        "q_proj",
        "k_proj",
        "v_proj",
        "o_proj",
        "up_proj",
        "gate_proj",
        "down_proj",
    ],
    "use_rslora": True,
}

DPO_CONFIG = {
    "num_train_epochs": 2,
    "logging_steps": 25,
    "eval_steps": 500,
    "save_steps": 100,
    "save_total_limit": 5,
    "per_device_train_batch_size": 1,
    "per_device_eval_batch_size": 1,
    "gradient_accumulation_steps": 16,
    "learning_rate": 5e-6,
    "max_length": 8192, # max length in the history chat + latest assistant response.
    "max_prompt_length": 6656, # max length in the history chat: user-assistant-...-assistant-user.
    "gradient_checkpointing": True,
    "weight_decay": 0.001,
    "optim": "rmsprop",
    "evaluation_strategy": "steps",
    "lr_scheduler_type": "cosine",
    "bf16": True,
}

Speeds, Sizes, Times

Below are some useful parameters showing the results of the latest training logs.

latest_logs = {'loss': 0.3716, 'grad_norm': 4.989994049072266, 'learning_rate': 1.0380020311950844e-10, 'rewards/chosen': 0.534086287021637, 'rewards/rejected': -0.6236276030540466, 'rewards/accuracies': 0.8899999856948853, 'rewards/margins': 1.1577140092849731, 'logps/rejected': -218.88198852539062, 'logps/chosen': -250.0700225830078, 'logits/rejected': -1.6214849948883057, 'logits/chosen': -1.9585875272750854, 'epoch': 1.99}

final_training_results = {'train_runtime': 30825.7138, 'train_samples_per_second': 1.432, 'train_steps_per_second': 0.089, 'train_loss': 0.483570138469306, 'epoch': 2.0}

As can be seen in the time used, in eight and a half hours we have managed to improve a state-of-the-art model, with very little hardware, in tasks adapted to Spanish. This can be seen in more detail in the following sections.

Evaluation

Testing Data, Factors & Metrics

Testing Data

To assess the performance of Large Language Models (LLMs), we have developed and utilized several high-quality corpora tailored to specific evaluation needs:

IIC/AQuAS: A manually curated corpus created by two computational linguists to evaluate language models in the task of Abstractive Question Answering in Spanish. It includes examples from domains such as finance, insurance, healthcare, law, and music.
IIC/RagQuAS. Another manually curated corpus developed by the same linguists to evaluate full RAG systems and language models in Abstractive Question Answering tasks in Spanish. This corpus spans a wide range of domains, including hobbies, linguistics, pets, health, astronomy, customer service, cars, daily life, documentation, energy, skiing, fraud, gastronomy, languages, games, nail care, music, skating, first aid, recipes, recycling, complaints, insurance, tennis, transportation, tourism, veterinary, travel, and yoga.
CAM: Designed for all CAM tasks, this corpus consists of frequently asked questions (FAQs) sourced from consumer-related topics on the websites of the Comunidad de Madrid. The questions are categorized into three levels of degradation—E1, E2, and E3—intended to measure the LLMs’ ability to understand and effectively respond to poorly formulated queries caused by spelling errors, varying levels of colloquialism, and similar issues. This task also falls under the Abstractive Question Answering category.
Shops: A multi-turn conversational corpus centered on policies from various clothing companies. The task involves Multi-turn Abstractive Question Answering.
Insurance: Another multi-turn conversational corpus, this one focuses on policies from various insurance companies. It also involves Multi-turn Abstractive Question Answering.

Each corpus includes the following columns: question, answer, and context(s) containing relevant information from which the model can derive the answer. In multi-turn tasks, a chat history is also provided.

The scoring process for LLMs involves measuring the similarity between the original answer and the one generated by the model. All corpora are private except for AQuAS and RagQuAS, which are publicly available and can serve as examples of the structure and content of the others.

Factors

These evaluations are very specific and do not encompass all the general scenarios to which the model could be exposed, since all evaluations are focused on solving tasks for RAG in very specific domains.

Metrics

The evaluation is based on using Llama-3.1-8B-Instruct to score the answers.

Results

Model	Average	AQuAS	RagQuAS	CAM	CAM_E1	CAM_E2	CAM_E3	Shops	Insurance
RigoChat-7b-v2	79.01	82.06	77.91	78.91	79.27	76.55	75.27	81.05	81.04
GPT-4o	78.26	85.23	77.91	78.00	74.91	73.45	77.09	78.60	80.89
stablelm-2-12b-chat	77.74	78.88	78.21	77.82	78.73	77.27	74.73	77.03	79.26
Mistral-Small-Instruct-2409	77.29	80.56	78.81	77.82	75.82	73.27	73.45	78.25	80.36
Qwen2.5-7B-Instruct	77.17	80.93	77.41	77.82	75.09	75.45	72.91	78.08	79.67
Meta-Llama-3.1-8B-Instruct	76.55	81.87	80.50	72.91	73.45	75.45	71.64	77.73	78.88
GPT-4o-mini	76.48	82.80	75.82	76.36	74.36	72.36	71.82	78.25	80.08
Phi-3.5-mini-instruct	76.38	81.68	81.09	75.82	74.73	71.45	70.36	77.43	78.45
gemma-2-9b-it	75.80	82.80	78.11	72.91	73.45	71.09	71.27	77.08	79.72
Ministral-8B-Instruct-2410	75.19	79.63	77.31	76.00	73.45	72.36	70.18	76.44	76.14
GPT-3.5-turbo-0125	74.78	80.93	73.53	76.73	72.55	72.18	69.09	75.63	77.64
Llama-2-7b-chat-hf	71.18	67.10	77.31	71.45	70.36	70.73	68.55	72.07	71.90
granite-3.0-8b-instruct	71.08	73.08	72.44	72.36	71.82	69.09	66.18	69.97	73.73
RigoChat-7b-v1	62.13	72.34	67.46	61.27	59.45	57.45	57.64	62.10	59.34
salamandra-7b-instruct	61.96	63.74	60.70	64.91	63.27	62.36	60.55	59.94	60.23

Summary

RigoChat-7b-v2 manages to significantly improve performance compared to Qwen-2.5 in the tasks for which it has been indirectly designed. On the other hand, it manages to outperform most state-of-the-art models in these tasks, demonstrating that with few resources LLMs can be aligned for specific use cases.

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Hardware Type: [More Information Needed]
Hours used: [More Information Needed]
Cloud Provider: [More Information Needed]
Compute Region: [More Information Needed]
Carbon Emitted: [More Information Needed]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

GPU NVIDIA A100 with Tensor Core.

Software

Ubuntu 22.04.5 LTS and the following requirements:

python=3.11
flash_attn>=2.5.8
datasets
numpy
trl
peft
huggingface_hub

Citation

@misc {instituto_de_ingeniería_del_conocimiento_2025,
    author       = { {Instituto de Ingeniería del Conocimiento} },
    title        = { RigoChat-7b-v2 },
    year         = 2025,
    url          = { https://huggingface.co/IIC/RigoChat-7b-v2 },
    doi          = { 10.57967/hf/4158 },
    publisher    = { Hugging Face }
}

Disclaimer

Please note that the model may contain biases or other undesired distortions based on the training data. When third parties implement systems or provide services based on this model, or use the model themselves, they are responsible for mitigating associated risks and ensuring compliance with applicable regulations, including those governing the use of artificial intelligence. The IIC, as the owner and creator of the model, shall not be held liable for any outcomes resulting from third-party use.

IIC
/

RigoChat-7b-v2

Model Card for RigoChat-7b-v2

Model Details

Model Description

Model Sources

Uses

Direct Use

Out-of-Scope Use

Bias, Risks, and Limitations

Recommendations

How to Get Started with the Model

To load the model and tokenizer

Sample generation

Tool Use

Training Details

Training Data

Training Procedure

Training Hyperparameters

Speeds, Sizes, Times

Evaluation

Testing Data, Factors & Metrics

Testing Data

Factors

Metrics

Results

Summary

Environmental Impact

Model Architecture and Objective

Compute Infrastructure

Hardware

Software

Citation

Disclaimer

Model Card Contact

Authorship Notice

Additional Information

Model tree for IIC/RigoChat-7b-v2