TimesFM 3.0 Forecasting

Tools

• timesfm_3_forecasting_run_multivariate Jointly forecast several columns of a CSV, optionally conditioned on covariates. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool. Returns: An interactive plot, a per-series accuracy table (backtest only), and the forecast values.
• timesfm_3_forecasting_run_rolling Backtest one column over many forecast origins instead of a single split. A single holdout is one draw and says little about skill. This walks the cut point backwards through the series, forecasts each origin blind, and reports the spread — including how often the model actually beats a last-value naive baseline. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool. Returns: A plot, a verdict summary, aggregate metrics across origins, and the per-origin table.
• timesfm_3_forecasting_run_univariate Forecast one numeric column of a CSV with TimesFM 3.0. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool. Returns: An interactive plot, an accuracy table (backtest only), and the forecast values.

No tool matches.

As the server's own listing gives them. We never call a tool.