Comparar métodos

	Sundial: Modelos Fundacionais Generativos de Séries Temporais	Chronos: Um Modelo de Fundação Tokenizado para Previsão de Séries Temporais	TimesFM
Área	Aprendizado profundo	Aprendizado profundo	Aprendizado profundo
Família	Machine learning	Machine learning	Machine learning
Ano de origem	2025	2024	2024
Autor original	Yong Liu et al. (Tsinghua)	Abdul Fatir Ansari et al. (Amazon)	Abhimanyu Das et al. (Google)
Tipo	Generative time-series foundation model family	Pre-trained language-model-based time-series forecaster	Pre-trained decoder-only transformer for zero-shot time-series forecasting
Fonte seminal	Liu, Y., Qin, G., Shi, X., Hu, T., Wang, J., & Long, M. (2025). Sundial: A family of highly capable time series foundation models. ICML.	Ansari, A. F., Stella, L., Turkmen, C., Zhang, X., Mercado, P., Shen, H., et al. (2024). Chronos: Learning the language of time series. Transactions on Machine Learning Research.	Das, A., Kong, W., Sen, R., & Zhou, Y. (2024). A decoder-only foundation model for time-series forecasting. ICML.
Outros nomes	Sundial TSF, Time-Series Foundation Model (Generative), Sundial ICML 2025, Zaman Serisi Temel Modeli (Sundial)	Chronos Forecasting Model, Amazon Chronos, Tokenized Time-Series LLM, Kronos Zaman Serisi Modeli	Time-series Foundation Model, Google TimesFM, TimesFM forecaster, Zaman Serisi Temel Modeli
Relacionados	3	2	3
Resumo	Sundial is a family of generative time-series foundation models introduced by Yong Liu and colleagues at Tsinghua University (ICML 2025). Pre-trained on large and diverse time-series corpora, Sundial employs a decomposition-based architecture paired with a generative forecasting head to produce probabilistic multi-horizon forecasts. It represents a shift toward general-purpose, zero-shot-capable models for real-world temporal prediction tasks.	Chronos is a family of pre-trained probabilistic forecasting models introduced by Ansari et al. at Amazon in 2024. It adapts the language-model paradigm to time series by quantizing continuous values into discrete tokens, enabling a standard transformer to be trained on a large heterogeneous corpus of time-series data. The result is a zero-shot forecasting model that generalizes across domains without requiring dataset-specific retraining.	TimesFM is a pre-trained foundation model for univariate time-series forecasting introduced by Abhimanyu Das, Weihao Kong, Rajat Sen, and Yichen Zhou from Google in 2024. The model adopts a decoder-only transformer architecture, similar in spirit to large language models, and is trained on a large corpus of real-world and synthetic time-series data. Its central innovation is the ability to perform accurate zero-shot forecasting across diverse domains without task-specific fine-tuning.