{ "index": [ "2024-01-01", "2024-02-01", "2024-03-01" ], "target": [120.0, 126.0, 133.0], "hist_variables": { "temperature": [20.0, 21.0, 22.0] }, "future_variables": { "temperature": [20.0, 21.0, 22.0, 23.0] }, "future_variables_index": [ "2024-01-01", "2024-02-01", "2024-03-01", "2024-04-01" ], "static_variables": { "region_id": 42 } }

Data Schemas

Complete type definitions for all request and response objects.

InputSeries

A single time-series object passed in the series array.

targetarray[number]requiredThe time-series values to forecast from.

indexarray[string]requiredISO date strings matching target length.

hist_variablesobjectoptionalHistorical covariate columns. Keys are variable names, values are arrays same length as target.

future_variablesobjectoptionalFuture covariate columns. Keys are variable names, values are arrays of length horizon.

future_variables_indexarray[string]optionalISO date strings for future_variables, length = horizon.

static_variablesobjectoptionalStatic scalar covariates. Each key maps to a single value.

ForecastAPIRequest

Body for POST /forecast and POST /forecast-jobs.

seriesarray[object]requiredOne or more time-series objects.

[item]object

targetarray[number]requiredThe time-series values to forecast from.

indexarray[string]requiredISO date strings matching target length.

hist_variablesobjectoptionalHistorical covariate columns. Keys are variable names, values are arrays same length as target.

future_variablesobjectoptionalFuture covariate columns. Keys are variable names, values are arrays of length horizon.

future_variables_indexarray[string]optionalISO date strings for future_variables, length = horizon.

static_variablesobjectoptionalStatic scalar covariates. Each key maps to a single value.

horizonintegerrequiredNumber of steps to forecast. Range: 1–10000.

freqstring (enum)required5min · 15min · 30min · H · D · W · M · MS · Q · Y

contextintegeroptionalHistory steps to use. Default: min(model_max, series_length).

quantilesarray[number]optionalQuantiles to return. Default: [0.1, 0.5, 0.9].

model_sizestringoptionalSize variant for multi-size models (chronos, chronos-2). e.g. "base".

ForecastAPIResponse

Body returned by POST /forecast and GET /forecast-jobs/{id} when completed.

statusstring (enum)completed · failed · in_progress

modelstringModel ID used for inference.

seriesarray[array[object]]Outer array = series, inner array = items per series.

[i][j]object

predictionobject

meanarray[number]Point forecast.

"0.1"array[number]10th percentile.

"0.5"array[number]Median.

"0.9"array[number]90th percentile.

indexarray[string]ISO date strings for forecast horizon.

metricsobject | nullOptional evaluation metrics.

SeriesPrediction

A single prediction object within series[i][j].prediction.

meanarray[number]Point forecast.

"0.1"array[number]10th percentile.

"0.5"array[number]Median.

"0.9"array[number]90th percentile.

ModelInfo

A single model object returned by GET /models.

model_idstringModel identifier.

namestringHuman-readable name.

sizesarray[string]Available size variants.

max_contextintegerMax historical context steps.

max_horizonintegerMax forecast horizon steps.

supports_covariatesbooleanSupports hist/future covariates.