Github · GitHub Repository Radar
google-research timesfm
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Stars
30,632
Forks
2,921
Watchers: 30,632
Language
License: Apache License 2.0
Repository Radar Score
63 / 100
Growth
- 7d
- +0
- 30d
- +0
- %
- 0.0%
Not enough metric snapshots yet to chart growth for this repository.
Score breakdown
- popularity 82
- growth 0
- activity 100
- freshness 100
- community 80
Need help integrating this stack?
Our team builds with modern open-source stacks. Tell us what you are shipping.
Get a quote →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.
- (NEW!) TimesFM 3.0 Checkpoint:
google/timesfm-3.0-pytorch. - Checkpoints (up to 2.5): TimesFM Hugging Face Collection.
- Google Research blog (New blog post for TimesFM 3.0 coming soon!).
- TimesFM in Google 1P Products:
- BigQuery ML: Enterprise level SQL queries for scalability and reliability.
- Google Sheets: For your daily spreadsheet.
- Vertex Model Garden: Dockerized endpoint for agentic calling.
This open version is not an officially supported Google product.
Latest Model Version: TimesFM 3.0
Archived Model Versions:
- 2.5: relevant code under
src/timesfm. - 1.0 and 2.0: relevant code archived in the subdirectory
v1. You canpip install timesfm==1.3.0to install an older version of this package to load them.
TimesFM 3.0 is out!
TimesFM 3.0 introduces native multivariate time-series forecasting, flexible covariate support (both past-only and past-and-future covariates), superior zero-shot generalist capabilities, and top performance across all three major time-series foundation model benchmarks.
- Native Multivariate & Univariate Forecasting with Covariates: Seamlessly forecast multi-channel multivariate series as well as individual univariate series, with native support for past-only and past-and-future dynamic covariates without per-task tuning.
- Top Benchmark Performance:
- 🥇 fev-bench: Rank #1 overall across 100 diverse real-world forecasting tasks.
- 🥇 TIME Benchmark: Rank #1 overall across 50 domain datasets and 98 evaluation tasks.
- 🥇 GIFT-Eval: Rank #1 among all foundation models.
Important: The TimesFM source code in this repository is licensed under Apache-2.0, and model weights up to version 2.5 remain Apache-2.0. However, for the time being, TimesFM 3.0 pretrained weights are distributed under the separate
timesfm-non-commercial-license-v1.0license and are restricted to non-commercial, non-production use. Commercial or production use of the default pretrained weights is not permitted.
Updated PyPI to timesfm=2.0.2. See
Install.
Added fine-tuning example using HuggingFace Transformers + PEFT (LoRA) — see
timesfm-forecasting/examples/finetuning/.
Also added unit tests (tests/) and incorporated several community fixes.
Shoutout to @kashif and @darkpowerxo.
Huge shoutout to @borealBytes for adding the support for AGENTS! TimesFM SKILL.md is out.
Added back the covariate support through XReg for TimesFM 2.5.
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
frequencyindicator. - has a couple of new forecasting flags.
Since the Sept. 2025 launch, the following improvements have been completed for TimesFM 2.5:
- ✅ Flax version of the model for faster inference.
- ✅ Covariate support via XReg (see Oct. 2025 update).
- ✅ Documentation, examples, and agent skill (see
timesfm-forecasting/). - ✅ Fine-tuning example with LoRA via HuggingFace Transformers + PEFT (see
timesfm-forecasting/examples/finetuning/). - ✅ Unit tests for core layers, configs, and utilities (see
tests/).
# Install TimesFM with PyTorch
pip install timesfm[torch]-
Clone the repository:
git clone https://github.com/google-research/timesfm.git cd timesfm -
Create a virtual environment and install with PyTorch:
# Using uv uv venv source .venv/bin/activate # Install the package in editable mode with torch uv pip install -e .[torch]
Pass a batch of 1D NumPy arrays of different context lengths to forecast univariate time series:
import numpy as np
from timesfm3 import TimesFM3Evaluator, ModelConfig
# Initialize TimesFM 3.0
config = ModelConfig(
checkpoint_path="google/timesfm-3.0-pytorch",
per_core_batch_size=32,
device="cuda"
)
forecaster = TimesFM3Evaluator(config)
# Two univariate series of different lengths (100 and 72 steps)
ts1 = np.linspace(0, 1, 100).astype(np.float32)
ts2 = np.sin(np.linspace(0, 24, 72)).astype(np.float32)
# Generate forecast (point predictions + 9 quantiles: 0.1 to 0.9)
outputs = list(forecaster.predict_batch([ts1, ts2], horizon=12, return_quantiles=True, use_symmetric_averaging=False))
print("Series 1 forecast shape:", outputs[0].forecast.shape) # (12,)
print("Series 1 quantiles shape:", outputs[0].quantiles.shape) # (12, 9)
print("Series 2 forecast shape:", outputs[1].forecast.shape) # (12,)
print("Series 2 quantiles shape:", outputs[1].quantiles.shape) # (12, 9)Pass a 2D array of shape (num_variates, context_length) along with optional
past-only and past-and-future covariates:
import numpy as np
from timesfm3 import TimesFM3Evaluator, ModelConfig
# Initialize TimesFM 3.0
config = ModelConfig(
checkpoint_path="google/timesfm-3.0-pytorch",
per_core_batch_size=16,
device="cuda"
)
forecaster = TimesFM3Evaluator(config)
context_len = 128
horizon = 24
# 3 target variates across past context: (3, 128)
target = np.random.randn(3, context_len).astype(np.float32)
# 1 past-only covariate channel across past context: (1, 128)
past_only_cov = np.random.randn(1, context_len).astype(np.float32)
# 2 past-and-future covariate channels across context + horizon: (2, 152)
past_future_cov = np.random.randn(2, context_len + horizon).astype(np.float32)
# Generate joint forecast across all 3 target variates
outputs = list(
forecaster.predict_batch(
contexts=[target],
horizon=horizon,
past_only_covariates=[past_only_cov],
past_future_covariates=[past_future_cov],
return_quantiles=True,
use_symmetric_averaging=False,
)
)
print("Multivariate forecast shape:", outputs[0].forecast.shape) # (3, 24)
print("Multivariate quantiles shape:", outputs[0].quantiles.shape) # (3, 24, 9)
Repository Radar analysis
Deterministic insights derived from public metadata and our observations — not personal testing or reviews.
Why this repository is interesting
- High absolute popularity (30,632 stars) signals broad adoption.
- Maintained recently (last push 2 weeks ago).
Who should use it
- Developers working primarily with Python
Potential use cases
- Reference or evaluate Python open-source approaches in this domain
Strengths
- Recent repository activity
- README present in our index
- Declared license: Apache License 2.0
- Substantial fork count (2,921) suggests reuse and contribution interest
Limitations / considerations
- Insights are derived from public metadata and our observations — not a substitute for code review
What to watch
- Re-check last push, issues, and releases on GitHub before production adoption
Strong signals: Strong community interest · Active maintenance
Source: GitHub (public metadata) + Repository Radar analysis. We do not claim ownership of third-party repositories.