A lot of misconception around the utility of foundational time series models on HN.
Sure, you could train a XGboost model on a time series with holiday or event covariates, but what if you have 10k different series to forecast for? You could train 10k different models, one for each series, but then they have very little history to learn from. Alternatively, you could train a single generic model to cover all 10k, but then you have to deal with scale mismatch between series in addition to the feature dev overhead between series. A TSFM saves a ton of work in modeling each of these series and could potentially generalize better to produce more accurate forecasts.
But why is the question, I run about 400k models every 30 minutes at work. I have tried all other approaches, but xgboost model per timeseries has never been beaten before. Even if I wasn't targetting accuracy, Xgboost loading a pretrained model and predicting with it and still faster.
Have you tried the latest TSFMs? Research (even in 2025) shows they outperform XGBoost models in certain scenarios. https://www.sciencedirect.com/science/article/pii/S266682702...
What's the input like? Does it incorporate language understanding, so that each time series has a natural language explanation of the series and all the data fields? If it takes in really descriptive text like that I can see the opportunity for generalization. If it's just a time series of non-descript numbers I see much much less opportunity for cross-series learning.
> 10k different series
who has these though? i cannot imagine somoone making sense of 10k timeseries.
All the businesses with product catalogs for example, every product can have several time series associated with it. If a business has 10k employees, you have, at minimum, 10k time series to perform employee analyst operations over. There are several hundred companies in the US with 10k+ employees. There are many more than 10k postal zipcodes in the US to perform forecasts over.
sorry i meant 10k types of timeseries data not individual items. I assume thats what gp meant.
The difference is the same, whether you have 10k series representing the same variables or 10k series representing different variables decomposes into the same forecasting problem that you can either over-generalize across all time-series, producing worse accuracy at the tails, or train a model for each individual forecast, producing worse accuracy in general due to limited data input.
As an energy retailer, I have 10k+ timeseries.
Though you could perhaps classify them first and use much fewer models.
No reason to spend time classifying and worrying about cross-series validation when you use a TSFM.
I forecast 400k timeseries every 30minutes at work.
What kind of infrastructure do you use for that scale? Are these computationally light ARIMA models?
We're running 150k models per day and while I'm sure I could scale this to running them every 30 minutes, it would mostly be me throwing money at the problem to parallelize the compute.