Skip to main content
DFlow supports spot crypto and prediction markets. It uses one trading model for every asset type. What changes is what you trade, not how you trade it. Every trade follows the same shape: request a quote, sign a transaction, submit it. Routes are planned at quote time and may be refined at execution by JIT routing, reflecting current onchain conditions.

​
Trading spot crypto

When trading spot crypto, users swap one SPL token for another.
  • A quote is requested.
  • The user signs and submits a transaction.
  • The trade routes through onchain liquidity venues.
  • The output token lands in the user’s wallet.
This covers SOL, stablecoins, and other SPL tokens, so builders can ship familiar swap experiences, trading UIs, and automated strategies without building execution logic from scratch.

​
Trading prediction markets

When trading prediction markets, users trade outcome tokens that represent possible results of an event.
  • A quote is requested for an outcome token.
  • The user signs and submits a transaction.
  • The trade executes through Kalshi’s prediction market liquidity.
  • The outcome token lands in the user’s wallet.
Prediction market trades execute through Concurrent Liquidity Programs (CLPs) as async, multi-transaction trades. Outcome tokens can be traded, and redeemed after a market closes, which opens up secondary trading, portfolio tracking, and post-resolution settlement flows.

​
Same model, different assets

From a builder’s perspective, both asset types follow the same flow:
  • Quotes are requested the same way.
  • Users sign and submit transactions the same way.
  • The same routing and execution primitives apply.
The differences come from asset behavior and liquidity structure, not the trading model:
  • Spot crypto trades complete atomically.
  • Prediction market trades always execute asynchronously through CLPs.
  • Prediction markets eventually resolve, so outcome tokens can be redeemed.
You interact with both through the same APIs.