/ Модель

Характеристики

Контекст
4 096 токенов
Параметры
1,5 млрд
Лицензия
Apache 2.0
Архитектура
qwen3 · dense
Базовая модель
Tucano2-qwen-1.5B-Base
Библиотека
transformers
Опубликована
5 фев 2026

Tucano2-qwen-1.5B-Instruct

Генерация текстаТекст → текст6 профессий

Инструкционная модель Qwen3 с 1,5 млрд параметров, дообученная для диалогов и генерации текста преимущественно на португальском языке.

1 054скачиваний за 30 дней2лайкавеса обновлены13 авг 2026

Бенчмарки

Как модель справилась с задачами пользователей ailist и что о ней публикуют открытые лидерборды

Тесты пользователейпока нет

Никто ещё не опубликовал тест этой модели. Соберите бенчмарк из своих задач — результат появится здесь.

  1. Добавьте модель в бенчмарк

    Вместе с другими моделями категории «Текст» — сравнение идёт внутри одной категории.

  2. Опишите свои задачи

    Напишите запросы, при необходимости приложите файлы и эталонные ответы.

  3. Запустите и опубликуйте

    Модели ответят на каждую задачу, судья оценит ответы. Опубликованный результат появится на этой странице.

Публичные оценки

Из открытых лидербордов. Сравнивайте результаты одного теста при одинаковых условиях — отсутствие оценки не значит, что модель хуже.

В подключённых источниках результатов для этой модели пока нет.

Что умеет модель

Описание составлено по карточке модели и метаданным Hugging Face

Tucano2-qwen-1.5B-Instruct — португалоязычная инструкционная модель компании Polygl0t, созданная на базе Tucano2-qwen-1.5B-Base. Это плотная Transformer-модель семейства Qwen3 с 1,5 млрд параметров и контекстом 4096 токенов; обучение включало один этап контролируемой настройки инструкций (SFT) и один этап оптимизации предпочтений (APO). Модель предназначена для исследований и разработки систем генерации текста на португальском языке. Она подходит для диалогов, суммаризации, генерации структурированных ответов, работы в сценариях RAG и использования инструментов или вызова функций.

Карточка сообщает о сравнительно сильных результатах семейства Tucano2 на португалоязычных тестах, но приведённые таблицы относятся к варианту 3.7B и не подтверждают напрямую качество версии 1.5B. Модель может выдавать фактически неверные или предвзятые ответы, токсичный текст, повторяться и быть многословной; другие языки, кроме португальского, могут обрабатываться хуже. Лицензия Apache 2.0 разрешает использование и адаптацию на её условиях; для реального применения рекомендуется отдельно оценивать риски и смещения.

Сценарии применения

  • ответы на вопросы на португальском языке
  • суммаризация португалоязычных текстов
  • генерация структурированных ответов
  • подготовка ответов по документам в сценарии RAG
  • вызов функций и применение инструментов в диалогах

Ограничения

Модель может галлюцинировать, воспроизводить предвзятость и токсичный текст, повторяться или давать многословные ответы; кроме португальского, другие языки могут обрабатываться хуже. Приведённые сравнительные оценки относятся к версии 3.7B, поэтому не характеризуют напрямую качество варианта 1.5B.

Для каких профессий подходит6

Разделы каталога профессий, в задачах которых используется эта модель

Все профессии

Также подходит для

Карточка модели

README от автора на Hugging Face, как он его опубликовал

Открыть на Hugging Face

Tucano2-qwen-1.5B-Instruct

Model Summary

Tucano2-qwen-1.5B-Instruct is an instruction-tuned Portuguese language model on top of Tucano2-qwen-1.5B-Base. It has been trained using a combination of one round of supervised fine-tuning (SFT) and one round of Anchored Preference Optimization (APO).

Despite its compact size, the model delivers strong performance across a wide range of Portuguese benchmarks and supports tasks such as retrieval-augmented generation, function calling and tool use, summarization, and structured output generation, among others.

All datasets, source code, and training recipes used to develop the Tucano2 series are fully open and reproducible.

Details

This repository has the source code used to train this model. The full configuration used for training is available in the following config files:

SFT Loss Curve

APO Rewards

Intended Uses

The primary intended use Tucano2-qwen-1.5B-Instruct is to serve as foundations for research and development involving Portuguese language modeling. You may also fine-tune and adapt Tucano2-qwen-1.5B-Instruct for deployment if your use follows the Apache 2.0 license. If you decide to use Tucano2-qwen-1.5B-Instruct as a basis for your fine-tuned model, please conduct your own risk and bias assessment.

Basic usage

from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
import torch

# Load model and tokenizer
model_id = "Polygl0t/Tucano2-qwen-1.5B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto"
)

# Configure generation parameters
generation_config = GenerationConfig(
    do_sample=True,
    temperature=0.1,
    top_k=50,
    top_p=1.0,
    repetition_penalty=1.2,
    max_new_tokens=150,
    pad_token_id=tokenizer.eos_token_id,
)

# Prepare chat messages
messages = [
    {"role": "user", "content": "Qual é a capital de Portugal?"}
]

# Apply chat template and generate
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

with torch.no_grad():
    outputs = model.generate(**inputs, generation_config=generation_config)

# Decode and print response
response = tokenizer.decode(outputs[0][len(inputs.input_ids[0]):], skip_special_tokens=True)
print(f"🤖 {response}")

Limitations

Like almost all other language models trained on large text datasets scraped from the web, the Tucano2-qwen-1.5B-Instruct shows behavior that does not make it an out-of-the-box solution to many real-world applications, especially those requiring factual, reliable, and nontoxic text generation. Tucano2-qwen-1.5B-Instruct is subject to the following:

  • Hallucinations: Tucano2-qwen-1.5B-Instruct can produce content that can be mistaken as true facts, but are misleading or entirely false, i.e., hallucination.

  • Biases and Toxicity: Tucano2-qwen-1.5B-Instruct inherits the social and historical stereotypes from the data used to train it. Given these biases, the model can produce toxic content, i.e., harmful, offensive, or detrimental to individuals, groups, or communities.

  • Language Limitations: Tucano2-qwen-1.5B-Instruct is primarily designed to interact with Portuguese. Other languages might challenge its comprehension, leading to potential misinterpretations or errors in response.

  • Repetition and Verbosity: Tucano2-qwen-1.5B-Instruct may get stuck on repetition loops (especially if the repetition penalty during generations is set to a meager value) or produce verbose responses unrelated to the prompt it was given.

Hence, even though Tucano2-qwen-1.5B-Instruct is released with a permissive license, we urge users to perform their risk analysis on them if they intend to use them for real-world applications.

Evaluations

The table below compares the Tucano2 (Instruct variant) series against other chat models of similar size. We divide our evaluations into three sets:

  • Knowledge & Reasoning: ARC-Challenge, ENEM, BLUEX, OAB Exams, BELEBELE, MMLU, GSM8K-PT
  • Instruction Following: IFEval-PT
  • Coding: HumanEval

The NPM (Normalized Performance Metric) provides a balanced view of model performance across tasks, accounting for each task's inherent difficulty by normalizing its evaluation score relative to its random baseline.

Total Avg.Knowledge & Reasoning (NPM)Instruction FollowingCoding
Tucano2-qwen-3.7B-Instruct53.6456.2241.6747.56
Jurema-7B53.0350.664775.61
Qwen2.5-3B-Instruct51.7147.3463.3370.73
Qwen3-4B51.3642.3379.3386.59
Gemma-3-Gaia-PT-BR-4b-it49.934570.3364.02
SmolLM3-3B49.5443.9969.6768.29
Llama-3.2-3B-Instruct45.8243.0862.6748.17
Qwen2.5-1.5B-Instruct41.3940.254248.78
Tucano2-qwen-1.5B-Instruct37.5439.6134.3326.22
Qwen3-1.7B36.328.246564.02
Tucano2-qwen-0.5B-Instruct26.0827.773010.37
Qwen3-0.6B22.2115.135539.02
Llama-3.2-1B-Instruct20.1415.3744.3329.27
Qwen2.5-0.5B-Instruct17.814.983124.39
Tucano-2b4-Instruct3.782.71150
Tucano-1b1-Instruct2.591.4213.330
TeenyTinyLlama-460m-Chat0.07-1.6812.330
Benchmarkn-shotTypeBaselineMetric
Knowledge & Reasoning
ARC-Challenge5-shotMC-Q&A25acc_norm
ENEM3-shotMC-Q&A20acc
BLUEX3-shotMC-Q&A22.5acc
OAB Exams3-shotMC-Q&A25acc
BELEBELE5-shotMC-Q&A25acc_norm
MMLU5-shotMC-Q&A25acc
GSM8K-PT0-shotMath Problems0flexible-extract
Instruction Following
IFEval-PT0-shotInstruction0prompt_level_loose_acc
Coding
HumanEval0-shotCoding0pass@1
BLUEXENEMOABARC ChallengeBELEBELEMMLUIFEval-PTGSM8K-PTHumanEval
Tucano2-qwen-3.7B-Instruct64.5372.9254.3160.3485.2264.6441.6753.8147.56
Jurema-7B63.4270.9664.9752.5688.4449.914730.2975.61
Qwen2.5-3B-Instruct56.8868.6546.7941.718458.2263.3351.970.73
Qwen3-4B63.2872.1550.343.0883.6726.9379.3339.8886.59
Gemma-3-Gaia-PT-BR-4b-it50.964.5243.4654.778.8951.4970.3351.2964.02
SmolLM3-3B53.5564.734152.7478.6753.2369.6744.4468.29
Llama-3.2-3B-Instruct52.0259.1344.9743.9378.5652.1462.6755.148.17
Qwen2.5-1.5B-Instruct52.0261.7944.2839.747651.914242.1648.78
Tucano2-qwen-1.5B-Instruct52.8562.743.4250.2677.5652.5434.3319.7126.22
Qwen3-1.7B50.4958.6432.6236.6764.8930.566526.9464.02
Tucano2-qwen-0.5B-Instruct40.3353.640.7338.6362.3341.463018.4910.37
Qwen3-0.6B34.9141.1529.7530.5137.1126.485531.6639.02
Llama-3.2-1B-Instruct30.0434.0130.8432.8241.5635.1544.3326.5629.27
Qwen2.5-0.5B-Instruct30.1834.0829.3427.4450.6739.543114.6924.39
Tucano-2b4-Instruct25.8720.0126.7431.972426.72152.050
Tucano-1b1-Instruct22.9519.9425.333024.8925.5913.331.670
TeenyTinyLlama-460m-Chat17.2518.1919.7326.8422.8926.9712.332.050

Math

Prompt:

<|im_start|>user
Como eu posso resolver o seguinte problema: 2x + 3 = 11?<|im_end|>
<|im_start|>assistant

Response:

Para encontrar a solução para esta equação, você precisa isolar x. Primeiro subtraia 3 de ambos os lados da equação e depois divida por 2. A resposta é x=4. Para verificar sua resposta substitua 4 na equação original (2 * 4) + 3 = 11, que simplifica para 8 + 3 = 11. Portanto, a solução está correta .<|im_end|>

Structured Output

Prompt:

<|im_start|>system
Sempre responda em formato JSON estruturado conforme solicitado.<|im_end|>
<|im_start|>user
Extraia e resuma as informações principais do e-mail abaixo. Escreva sua resposta (em formato JSON) contendo as seguintes chaves: 'assunto', 'remetente', 'destinatário', 'resumo'. Aqui está o e-mail:

Assunto: RE: Atualização sobre o Projeto de Conservação de Bacias Hidrográfica

Carlos,

Obrigado por enviar o currículo atualizado. Tive a oportunidade de revisá-lo e achei que ficou ótimo! Você fez um excelente trabalho ao tornar as informações acessíveis e envolventes para uma ampla audiência.

Em relação à possível resistência dos membros da comunidade, concordo que devemos estar preparados para abordar suas preocupações. Uma ideia que tive foi incluir alguns exemplos práticos de como o uso indevido das águas no rio Amazonas ou na Baía de Guanabara e o descarte incorreto de resíduos sólidos têm prejudicado essas áreas. Ouvi-los falar diretamente sobre as consequências pode ser muito mais persuasivo do que simplesmente mostrar dados estatísticos.

Além disso, também considero importante destacarmos os ganhos econômicos trazidos pela adoção de práticas mais sustentáveis, como tarifas de água menores e gastos diminutos relacionados à limpeza urbana. No final das contas, as pessoas costumam responder bem quando veem vantagens tangíveis além dos aspectos puramente ambientais.

Diga-me o que você pensa dessas sugestões! Ficarei contente em ajudar a desenvolver ainda mais estratégias para vencer essa resistência e engajar as pessoas nessa nossa jornada.

Atenciosamente,
Helena<|im_end|>
<|im_start|>assistant

Response:

```json

{
  "assunto": "RE: Atualização sobre o Projeto de Conservação de Bacias Hidrográficas",
  "remetente": "Helena",
  "destinatario": "Carlos",
  "resumo": "A remetente elogia o novo currículo enviado pelo destinatário, sugerindo adicionar casos reais impactantes nas bacias hidrográficas brasileiras para aumentar a eficácia educacional. Ela enfatiza a importância de evidenciar benefícios financeiros associados às práticas sustentáveis."
}

```<|im_end|>

Function Call / Tool Use

Prompt:

<|im_start|>system
Utilize as ferramentas para resolver as tarefas que o usuário lhe solicitar. Se as ferramentas forem insuficientes, explique ao usuário que não é possível completar a tarefa.

\# Tools / Ferramentas

Você pode chamar uma ou mais funções para auxiliar na consulta do usuário.

Você recebe assinaturas de funções dentro de tags XML <tools></tools>:
<tools>
{"type": "function", "function": {"name": "create_todo", "description": "Cria uma nova tarefa no aplicativo de todo.", "parameters": {"type": "object", "properties": {"task_description": {"type": "string", "description": "Descrição da tarefa a ser criada."}, "due_date": {"type": "string", "description": "Data de entrega da tarefa no formato YYYY-MM-DD."}}, "required": ["task_description", "due_date"]}}}
</tools>

Para cada chamada de função, retorne um objeto json com o nome da função e os argumentos dentro das tags XML <tool_call></tool_call>:
<tool_call>
{"name": <function-name>, "arguments": <args-json-object>}
</tool_call><|im_end|>
<|im_start|>user
Eu preciso criar uma nova tarefa para o meu projeto.<|im_end|>
<|im_start|>assistant
Claro, eu posso ajudar com isso. Você poderia, por favor, me fornecer a descrição da tarefa e a data de entrega?<|im_end|>
<|im_start|>user
A tarefa é finalizar o relatório do projeto e a data de entrega é 2022-03-15.<|im_end|>
<|im_start|>assistant

Response:

<tool_call>
{"name": "create_todo", "arguments": {"task_description": "Finalizing the project report", "due_date": "2022-03-15"}}
</tool_call><|im_end|>

Similarity Scoring

Prompt:

<|im_start|>user
Frase 1: O artista pintou um quadro inspirado nos campos floridos da primavera.
Frase 2: O criador produziu uma obra que representava paisagens urbanas cinzentas.
Pergunta: Qu\u00e3o similares s\u00e3o as duas frases? D\u00ea uma pontua\u00e7\u00e3o entre 1,0 a 5,0.
Resposta:<|im_end|>
<|im_start|>assistant

Response:

2.5. As duas frases apresentam situações diferentes quanto à temática e estilo artístico, resultando numa pontuação média de similaridade de 2.5, indicando pouca relação entre elas.<|im_end|>

Classification

Prompt:

<|im_start|>user
Classifique o sentimento da seguinte resenha de filme como 'Positiva', 'Negativa' ou 'Neutra':

"O filme foi uma experiência cinematográfica incrível, com atuações excepcionais e uma trama envolvente que me manteve na ponta da cadeira do começo ao fim."<|im_end|>
<|im_start|>assistant

Response:

Sentimento: Positivo. A crítica destaca aspectos positivos como performances excelentes e um enredo cativante, indicando uma opinião favorável sobre o filme.<|im_end|>

Below, we compare the performance of Tucano2-qwen-1.5B-Instruct with Qwen3-1.7B, a strong baseline in the 1.5B parameter range. The percentages represent the absolute difference in performance between the two models on each benchmark. All other plots can be found in the .plots folder.

Tucano2-qwen-1.5B-Instruct vs Qwen3-1.7B

Performance Comparison

Cite as 🤗

@misc{correa2026tucano2cool,
      title={{Tucano 2 Cool: Better Open Source LLMs for Portuguese}}, 
      author={Nicholas Kluge Corr{\^e}a and Aniket Sen and Shiza Fatimah and Sophia Falk and Lennard Landgraf and Julia Kastner and Lucie Flek},
      year={2026},
      eprint={2603.03543},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2603.03543}, 
}

Aknowlegments

Polyglot is a project funded by the Federal Ministry of Education and Research (BMBF) and the Ministry of Culture and Science of the State of North Rhine-Westphalia (MWK) as part of TRA Sustainable Futures (University of Bonn) and the Excellence Strategy of the federal and state governments.

We also gratefully acknowledge the granted access to the Marvin cluster hosted by University of Bonn along with the support provided by its High Performance Computing & Analytics Lab.

License

Tucano2-qwen-1.5B-Instruct is licensed under the Apache License, Version 2.0. For more details, see the LICENSE file.

Как запустить

На своей GPU в ailist. Цена — по текущим предложениям провайдеров

GPU в ailist

Универсальный рантайм · transformers

Можно запустить

Подбираем GPU…

Класс архитектуры
Qwen3ForCausalLM
Ревизия весов
82b7b2b

0 моделей