Skip to main content
POST
/
forecast
Request Time Series Forecast
curl --request POST \
  --url https://api.nolano.ai/forecast \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "series": [
    {
      "timestamps": [
        "2023-01-01T00:00:00",
        "2023-01-02T00:00:00",
        "2023-01-03T00:00:00"
      ],
      "values": [
        100.5,
        102.3,
        98.7
      ]
    }
  ],
  "forecast_horizon": 12,
  "data_frequency": "Daily",
  "forecast_frequency": "Daily",
  "confidence": 0.95
}
'
{
  "forecast_timestamps": [
    "2024-01-01T00:00:00",
    "2024-01-02T00:00:00"
  ],
  "lower_bound": [
    145.2,
    146.8
  ],
  "median": [
    150,
    151.5
  ],
  "upper_bound": [
    154.8,
    156.2
  ]
}
Generate predictions for your time series data using state-of-the-art foundation models. This endpoint supports univariate forecasting with multiple model options optimized for different use cases.
Processing Time: Forecasts typically complete in 2-10 seconds depending on data size and model complexity.

​
Available Models

Choose the best model for your forecasting needs. Each model is optimized for specific use cases and data characteristics:
Model IDDescriptionBest For
forecast-model-1General-purpose foundation model (TOTO)Most time series patterns
forecast-model-2Trend-focused modelData with strong trends
forecast-model-3Seasonal modelSeasonal patterns
forecast-model-4Volatility modelHigh-variance data
For detailed model information and performance characteristics, see our Supported Models page.

​
Example Requests

curl --location 'https://api.nolano.ai/forecast' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer ak_your_api_key_here' \
--header 'X-Model-Id: forecast-model-1' \
--data '{
  "series": [
    {
      "timestamps": [
        "2023-01-01T00:00:00",
        "2023-01-02T00:00:00",
        "2023-01-03T00:00:00",
        "2023-01-04T00:00:00",
        "2023-01-05T00:00:00"
      ],
      "values": [100, 102, 98, 105, 103]
    }
  ],
  "forecast_horizon": 7,
  "data_frequency": "Daily",
  "forecast_frequency": "Daily",
  "confidence": 0.95
}'

​
Response Format

The API returns forecast data with prediction intervals:
Example Response
{
  "forecast_timestamps": [
    "2024-01-01T00:00:00",
    "2024-01-02T00:00:00", 
    "2024-01-03T00:00:00"
  ],
  "lower_bound": [145.2, 146.8, 148.1],
  "median": [150.0, 151.5, 153.2],
  "upper_bound": [154.8, 156.2, 158.3]
}

​
Data Requirements

  • Minimum data points: 10 historical observations
  • Maximum forecast horizon: 100 periods
  • Supported frequencies: Seconds, Minutes, Hours, Daily, Weekly, Monthly, Quarterly, Yearly
  • Data format: Chronologically ordered timestamps with corresponding numerical values

​
Error Handling

The API returns structured error responses with specific error codes:
  • UNAUTHORIZED - Invalid or missing API key
  • INVALID_REQUEST - Validation failed for request parameters
  • DATA_VALIDATION_ERROR - Issues with time series data format
  • RATE_LIMIT_EXCEEDED - API rate limit exceeded
  • INTERNAL_ERROR - Unexpected server error
Always check the HTTP status code and parse the error object for detailed information about failures.

Authorizations

​
Authorization
string
header
required

API key authentication. Include your API key with 'Bearer' prefix.

Headers

​
X-Model-Id
enum<string>
default:forecast-model-1

Model ID to use for forecasting

Available options:
forecast-model-1,
forecast-model-2,
forecast-model-3,
forecast-model-4

Body

application/json
​
series
object[]
required

Array containing one time series object

Required array length: 1 element
​
forecast_horizon
integer
required

Number of future periods to predict

Required range: 1 <= x <= 100
Example:

12

​
data_frequency
enum<string>
required

Frequency of input timestamps

Available options:
Seconds,
Minutes,
Hours,
Daily,
Weekly,
Monthly,
Quarterly,
Yearly
Example:

"Daily"

​
forecast_frequency
enum<string>
required

Desired frequency for forecast output (must match data_frequency)

Available options:
Seconds,
Minutes,
Hours,
Daily,
Weekly,
Monthly,
Quarterly,
Yearly
Example:

"Daily"

​
confidence
number
default:0.95

Confidence level for prediction intervals

Required range: 0.1 <= x <= 0.99
Example:

0.95

Response

Forecast generated successfully

​
forecast_timestamps
string<date-time>[]
required

Timestamps for forecasted periods

Example:
[
  "2024-01-01T00:00:00",
  "2024-01-02T00:00:00"
]
​
lower_bound
number<float>[]
required

Lower bounds of prediction intervals

Example:
[145.2, 146.8]
​
median
number<float>[]
required

Point forecasts (median predictions)

Example:
[150, 151.5]
​
upper_bound
number<float>[]
required

Upper bounds of prediction intervals

Example:
[154.8, 156.2]