research
Timer-S1: 8.3B time series foundation model achieves state-of-the-art forecasting on GIFT-Eval
Researchers have introduced Timer-S1, a Mixture-of-Experts time series foundation model with 8.3 billion total parameters and 750 million activated parameters per token. The model achieves state-of-the-art forecasting performance on the GIFT-Eval leaderboard, with the best MASE and CRPS scores among pre-trained models.