Inference

Calibrated Brier-weighted ensemble forecast across multiple models

Produce an empirically calibrated ensemble forecast combining multiple LLM models weighted by their verified historical Brier score.

post/predict/ensemble

Request

  • Base URL: https://foresea.ink (the document declares no server; this is the origin it was published from)
  • URL: https://foresea.ink/predict/ensemble
  • Auth: none declared

Request body

questionstring required

Forecasting question

modelsstring[] nullable

Models to include in ensemble (defaults to top council models)

categorystring nullable

Optional domain category for calibration weights

market_probabilitynumber nullable

Reference market price if known

market_platformstring nullable

Reference market platform

member_forecastsobject[] nullable

Optional pre-computed member forecasts [{'model': '...', 'probability': 0.7, 'rationale': '...'}]

Response

Successful Response

questionstring required
ensemble_probabilitynumber required
confidence_intervalnumber[] required
dispersion_stdnumber required
consensus_levelstring required
edgenumber nullable
market_probabilitynumber nullable
member_contributionsobject[] required
rationale_summarystring required

Changes

No recorded changes to this operation across all 1 revision of this API.