
Request forecasts and market evidence from an agent while tracking separately metered API usage.
- Language Python
Find your toolkit
Evaluate an estimate by how it was produced and which event it describes. Forecasting tools and AI research can offer a probability estimate, but the supporting evidence determines what you can assess.
A forecast and the market price are different measures. Compare the evidence, assumptions and time horizon before interpreting a difference between them.

Request forecasts and market evidence from an agent while tracking separately metered API usage.

Source-linked event reports, model-price comparisons and developer access through API, MCP and CLI.

Draft an editable market rule, confirm a paper test and inspect observations, modeled costs and saved results.

Model-based market research with frozen Buy calls, public grading and wallet notifications.

Compare category-specific wallet records, inspect calibration and receive explained conviction alerts by email or Telegram.

A Kalshi-first research assistant with backtests, scheduled alerts and Polymarket comparisons; users execute orders themselves.

A signal wire and research assistant for public wallet activity, opposing evidence and resolution-clause checks.

A live AI forecasting benchmark with resolved-event scores, comparable-event standings and inspectable research traces.

A publication of political forecasting essays, model updates and prediction-market trading notes.

Weather-market research combining station observations, forecast-model comparisons, bias tools, alerts and tiered APIs.

Request an AI forecast from a link, question or image, then compare market odds and news context.

Make calls with friends using virtual Cred, private groups and imported odds for comparison.