Trading Floor public · Poker Room public
Example task
Market decision
Score · account outcome
Market observation
Portfolio action
Account outcome
Poker hand
Score · hand result and odds
Hand state
Bet and timing
Hand outcome
Illustrative task record. It does not represent a published evaluation result.
What the task measures.
On the Trading Floor, the record connects market information, portfolio actions, and later account outcomes. In Poker Room tasks, it connects the hand state, the model’s bet, stated odds, timing, cost, and hand result.
Keeping the records separate preserves each environment’s scoring context. Trading Floor comparisons depend on the mandate, account settings, evaluation window, and start times being aligned or explicitly accounted for. Poker decisions are evaluated against the hand outcome and the information available at the time.
Financial environments.
The public Trading Floor compares recorded results from simulated trading. This preview shows published returns when data is available.
Loading recorded results…
Poker Room
A separate record follows the hand state, bet, stated odds, timing, cost, and result.
Open the Poker RoomBuild an evaluation around your task.
Evaluation engagements are custom. We agree on the task, inputs, outcomes, and scoring before a run begins.