https://github.com/google-research/timesfm

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

Science Score: 36.0%

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
○
DOI references
✓
Academic publication links
Links to: arxiv.org
○
Committers with academic emails
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (11.5%) to scientific vocabulary

Keywords from Contributors

transformer tensors reinforcement-learning deep-neural-networks distributed xgboost speech-recognition large-language-models anomaly-detection data-mining

Last synced: 6 months ago · JSON representation

Repository

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

Basic Info

Host: GitHub
Owner: google-research
License: apache-2.0
Language: Python
Default Branch: master
Homepage: https://research.google/blog/a-decoder-only-foundation-model-for-time-series-forecasting/
Size: 2.29 MB

Statistics

Stars: 6,518
Watchers: 53
Forks: 569
Open Issues: 147
Releases: 2

Created almost 2 years ago · Last pushed 6 months ago

Metadata Files

Readme License

TimesFM

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

Paper: A decoder-only foundation model for time-series forecasting, ICML 2024.
All checkpoints: TimesFM Hugging Face Collection.
Google Research blog.
TimesFM in BigQuery: an official Google product.

This open version is not an officially supported Google product.

Latest Model Version: TimesFM 2.5

Archived Model Versions:

1.0 and 2.0: relevant code archived in the sub directory v1. You can pip install timesfm==1.3.0 to install an older version of this package to load them.

Update - Sept. 15, 2025

TimesFM 2.5 is out!

Comparing to TimesFM 2.0, this new 2.5 model:

uses 200M parameters, down from 500M.
supports up to 16k context length, up from 2048.
supports continuous quantile forecast up to 1k horizon via an optional 30M quantile head.
gets rid of the frequency indicator.
has a couple of new forecasting flags.

Along with the model upgrade we have also upgraded the inference API. This repo will be under construction over the next few weeks to

add support for an upcoming Flax version of the model (faster inference).
add back covariate support.
populate more docstrings, docs and notebook.

Install

Clone the repository: shell git clone https://github.com/google-research/timesfm.git cd timesfm
Create a virtual environment and install dependencies using uv: ```shell

Create a virtual environment

uv venv

Activate the environment

source .venv/bin/activate

Install the package in editable mode with torch

uv pip install -e .[torch]

Or with flax

uv pip install -e .[flax] ```
[Optional] Install your preferred torch / jax backend based on your OS and accelerators (CPU, GPU, TPU or Apple Silicon).:

Install PyTorch.
Install Jax for Flax.

Code Example

```python import torch import numpy as np import timesfm

torch.setfloat32matmul_precision("high")

model = timesfm.TimesFM2p5200Mtorch.frompretrained("google/timesfm-2.5-200m-pytorch")

model.compile( timesfm.ForecastConfig( maxcontext=1024, maxhorizon=256, normalizeinputs=True, usecontinuousquantilehead=True, forceflipinvariance=True, inferispositive=True, fixquantilecrossing=True, ) ) pointforecast, quantileforecast = model.forecast( horizon=12, inputs=[ np.linspace(0, 1, 100), np.sin(np.linspace(0, 20, 67)), ], # Two dummy inputs ) pointforecast.shape # (2, 12) quantileforecast.shape # (2, 12, 10): mean, then 10th to 90th quantiles. ```

Owner

Name: Google Research
Login: google-research
Kind: organization
Location: Earth

Website: https://research.google
Repositories: 226
Profile: https://github.com/google-research

Committers

Last synced: 6 months ago

All Time

Total Commits: 174
Total Committers: 22
Avg Commits per committer: 7.909
Development Distribution Score (DDS): 0.649

Past Year

Commits: 69
Committers: 10
Avg Commits per committer: 6.9
Development Distribution Score (DDS): 0.551

Top Committers

Name	Email	Commits
Rajat Sen	r**1@g**m	61
Yichen Zhou	y**u@g**m	29
tanmayshishodia	t**a@n**m	16
Justin Guese	g**n@g**m	15
misha-chertushkin	c**l@g**m	11
misha-chertushkin	c**n@g**m	8
Kashif Rasul	k**l@g**m	6
Killer3048	g**1@m**u	5
tanmayshishodia	t**a@g**m	5
Funto-Adeyemi	m**9@g**m	4
geetu040	r**0@g**m	2
periodLeo	l**n@p**e	2
Jan Lebert	s**c@u**m	1
Matthew Heidemann	m**n@g**m	1
Nan Huang	t**n@g**m	1
Sarthak Mishra	s**5@g**m	1
Sarthak Mishra	s**1@i**m	1
andreramosfdc	a**c@g**m	1
azul	f**z@g**m	1
dependabot[bot]	4**]@u**m	1
ram-from-tvl	r**l@g**m	1
zgeor	z**r@u**m	1

Committer Domains (Top 20 + Academic)

google.com: 2 in.ibm.com: 1 proton.me: 1 mail.ru: 1 nutanix.com: 1

Issues and Pull Requests

Last synced: 6 months ago

All Time

Total issues: 182
Total pull requests: 186
Average time to close issues: 12 days
Average time to close pull requests: 8 days
Total issue authors: 159
Total pull request authors: 36
Average comments per issue: 0.74
Average comments per pull request: 0.66
Merged pull requests: 100
Bot issues: 0
Bot pull requests: 29

Past Year

Issues: 88
Pull requests: 84
Average time to close issues: 6 days
Average time to close pull requests: 8 days
Issue authors: 74
Pull request authors: 19
Average comments per issue: 0.38
Average comments per pull request: 0.43
Merged pull requests: 42
Bot issues: 0
Bot pull requests: 19

View more stats

Top Authors

Issue Authors

Arsa-Nik (4)
jobs-git (4)
JackeyLee007 (3)
AzulGarza (3)
veelion (2)
algomaschine (2)
sdmorrey (2)
aleksmaksimovic (2)
NidhiRavishankar (2)
androidwoman (2)
sebassaras02 (2)
zhousheng89 (2)
Mhdaw (2)
forhonourlx (2)
dafmdev (2)

Pull Request Authors

rajatsen91 (56)
dependabot[bot] (29)
siriuz42 (17)
misha-chertushkin (11)
kashif (6)
AzulGarza (5)
faisal-fida (4)
periodLeo (4)
abdulfatir (4)
sarthak405 (4)
JustinGuese (3)
TeddyHuang-00 (3)
jobs-git (2)
heidmotron (2)
Killer3048 (2)

Top Labels

Issue Labels

Pull Request Labels

dependencies (29) python (13)

Packages

Total packages: 3
Total downloads:
- pypi 37,563 last-month

Total dependent packages: 0
(may contain duplicates)
Total dependent repositories: 0
(may contain duplicates)
Total versions: 42
Total maintainers: 4

proxy.golang.org: github.com/google-research/timesfm

Documentation: https://pkg.go.dev/github.com/google-research/timesfm#section-documentation
License: apache-2.0
Latest release: v1.2.6
published over 1 year ago

Versions: 2
Dependent Packages: 0
Dependent Repositories: 0

Rankings

Dependent packages count: 6.1%

Average: 6.3%

Dependent repos count: 6.5%

Last synced: 7 months ago

pypi.org: timecopilot-timesfm

Open weights time-series foundation model from Google Research.

Homepage: https://github.com/google-research/timesfm
Documentation: https://timecopilot-timesfm.readthedocs.io/
License: apache-2.0
Latest release: 0.2.1
published 6 months ago

Versions: 3
Dependent Packages: 0
Dependent Repositories: 0
Downloads: 1,353 Last month

Rankings

Dependent packages count: 8.8%

Average: 29.3%

Dependent repos count: 49.8%

Maintainers (1)

fgarzar

Last synced: 6 months ago

pypi.org: timesfm

Open weights time-series foundation model from Google Research.

Homepage: https://github.com/google-research/timesfm
Documentation: https://timesfm.readthedocs.io/
License: apache-2.0
Latest release: 1.3.0
published 9 months ago

Versions: 37
Dependent Packages: 0
Dependent Repositories: 0
Downloads: 36,210 Last month

Rankings

Dependent packages count: 10.9%

Average: 36.2%

Dependent repos count: 61.4%

Maintainers (3)

rajatsen91 JustinGuese siriuz42

Last synced: 6 months ago

Dependencies

environment.yml pypi

datasetsforecast *
einshape *
fire *
huggingface_hub *
jax *
neuralforecast *
orjson *
paxml *
praxis *
statsforecast *
utilsforecast *

pyproject.toml pypi

jax >=0.4.26
numpy >=1.26.4
pandas >=2.1.4
paxml >=1.4.0
praxis >=1.4.0

https://github.com/google-research/timesfm

Science Score: 36.0%

Keywords from Contributors

Repository

Basic Info

Statistics

Metadata Files

README.md

TimesFM

Update - Sept. 15, 2025

Install

Create a virtual environment

Activate the environment

Install the package in editable mode with torch

Or with flax

Code Example

Owner

Committers

All Time

Past Year

Top Committers

Committer Domains (Top 20 + Academic)

Issues and Pull Requests

All Time

Past Year

Top Authors

Issue Authors

Pull Request Authors

Top Labels

Issue Labels

Pull Request Labels

Packages

proxy.golang.org: github.com/google-research/timesfm

Rankings

pypi.org: timecopilot-timesfm

Rankings

Maintainers (1)

pypi.org: timesfm

Rankings

Maintainers (3)

Dependencies