Time Series Forecasting
Transformers
Safetensors
sundial
text-generation
time series
time-series
forecasting
foundation models
pretrained models
generative models
time series foundation models
custom_code
Instructions to use thuml/sundial-base-128m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use thuml/sundial-base-128m with Transformers:
# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("thuml/sundial-base-128m", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Exogenous variables and finetuning
#1
by iamshnoo - opened
The base model seems to perform really well on even unknown tasks! I was wondering if there are some examples of using exogenous variables apart from the time series data itself. And also, if there are examples for finetuning the model on our own data?
Thanks!
Great to hear that the base model is performing well for unknown tasks! Stay tuned for the update for finetuning—we’ll include notebooks and scripts to make this easy in the next few months!
Hi, if you’re interested in incorporating external variables or adapting a TSFM to your own data, you might find CoRA (https://arxiv.org/abs/2510.12681) to be a helpful and well-evaluated approach to explore.