Timesfm 3.0 Pytorch
Timesfm 3.0 Pytorch by google, a time-series-forecasting model. Understand and compare features, benchmarks, and capabilities.
Comparison
| Feature | Timesfm 3.0 Pytorch | Interfaze |
|---|---|---|
| Input Modalities | text | image, text, audio, video, document |
| Native OCR | No | Yes |
| Long Document Processing | No | Yes |
| Language Support | unknown | 162+ |
| Native Speech-to-Text | No | Yes |
| Native Object Detection | No | Yes |
| Guardrail Controls | No | Yes |
| Context Input Size | unknown | 1M |
| Tool Calling | No | Tool calling supported + built in browser, code execution and web search |
Scaling
| Feature | Timesfm 3.0 Pytorch | Interfaze |
|---|---|---|
| Scaling | Self-hosted/Provider-hosted with quantization | Unlimited |
View model card on Hugging Face
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
This repository contains the official PyTorch weights and configurations for TimesFM 3.0.
License
This model is released under the TimesFM Non-Commercial License v1.0.
Model Details
- Architecture: Stacked Mixing Transformer with Variate Attention and CPM Iterative RevIN.
- Context Patch Length: 32
- Forecast Horizon Patch Length: 64
- Layers: 20 transformer layers (model dim: 1280, heads: 16)
- Quantiles: (median at index 4)
Data
timesfm-3.0 is pretrained using
- GiftEvalPretrain excluding the datasets that overlap with fev-bench
- Wikipedia Pageviews, cutoff Nov 2023 (see paper for details).
- Google Trends top queries, cutoff EoY 2022 (see paper for details).
- Synthetic and augmented data.
Citation
@article{das2023decoder, title={A decoder-only foundation model for time-series forecasting}, author={Das, Abhimanyu and Kong, Weihao and Sen, Rajat and Zhou, Yichen}, journal={arXiv preprint arXiv:2310.10688}, year={2023} }