Skip to content

[GitHub Trending] google-research/timesfm

8 relevance
Score Breakdown
technical depth
9
novelty
8
actionability
7
community
7
strategic
8
personal
8

Scored daily by a customisable AI persona to surface the most relevant engineering leadership news.

Google's time-series foundation model, relevant to data engineering and ML infrastructure.

AI/ML github.com
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting. - google-research/timesfm
Summary

Google Research released TimesFM 2.5, a 200M parameter time-series foundation model (down from 500M) supporting up to 16k context length and continuous quantile forecasts via an optional 30M head. The open-source version integrates with BigQuery ML, Google Sheets, and Vertex Model Garden, and now includes Flax inference, XReg covariate support, LoRA fine-tuning via HuggingFace Transformers + PEFT, and an agent skill (SKILL.md). The model is available on PyPI with torch or Flax backends and can be compiled for production forecasting with configurable normalization, flip invariance, and quantile crossing fixes.

Author

google-research