Google's TimesFM: open-source time-series foundation model does zero-shot forecasting
mdancho84 · x · 2026-11-15
Google Research's TimesFM, a pretrained time-series foundation model (100B real-world data points), forecasts sales trends, market prices, traffic and energy demand zero-shot without fine-tuning — reportedly beating supervised models trained on your own data. The latest TimesFM 3.0 checkpoint is on Hugging Face, the repo has 35.5k GitHub stars, ships under Apache-2.0 and runs locally. Architecture is decoder-only with patch-based I/O (inputpatchlen=32, outputpatchlen=128), from the ICML 2024 paper; an official BigQuery integration also exists.
More from Models
- "Phonebook-trained models can't write Shakespeare": data distribution beats any harness — gerardsans · 2026-11-15
- Harness skeptic: model training-data support drives performance, not the harness — gerardsans · 2026-11-15
- Security veteran: traditional tooling is obsolete, run an agentic hacker against your own systems — Kyrannio · 2026-11-15
- AIFlow: China looks set to save local AI — AIFlow_ML · 2026-11-15
- All major AI models appear to have stopped using the em dash — mark_k · 2026-11-15
- Same 3D gummy orange prompt, two models: Fable 5.2 vs Opus 5.5 side by side — FinanceYF5 · 2026-11-15