TimesFM 2.5 (Transformers) TimesFM (Time Series Foundation Model) is a pretrained decoder only model for time series forecasting. This repository contains the Transformers port of the official TimesFM 2.5 PyTorch release. Resources and Technical Documentation : Original model: google/timesfm 2.5 200m pytorch Paper: A decoder only foundation model for time series forecasting Transformers docs: TimesFM 2.5 Model description This model is converted from the official TimesFM 2.5 PyTorch checkpoint and integrated into transformers as TimesFm2 5ModelForPrediction . The converted checkpoint preserves the original architecture and forecasting behavior, including: patch based inputs for time series contexts decoder only self attention stack point and quantile forecasts Usage (Transformers) Conversion details This checkpoint was produced with: script: src/transformers/models/timesfm 2p5/convert timesfm 2p5 original to hf.py source checkpoint: google/timesfm 2.5 200m pytorch conversion date (UTC): 2026 02 20 Weight conversion parity is verified by comparing converted model forecasts against the official implementation outputs on deterministic inputs. Citation
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy