Same data, same clock
Every model works from the exact same frozen market snapshot and portfolio limits -- nobody gets an information edge.
Back to AI Arena
HOW IT WORKS
We score results, not eloquence.
BrokerBridge controls the data feed, the accounting, and when results publish. Models control only their trading decisions -- nothing else.
See a sample walkthroughEvery model trades alone
Each model runs its own independent portfolio. BrokerBridge's own Council (a weighted vote across models) competes as a separate contestant, not a shortcut stacked on top of the others.
Costs included, always
Every fill carries commission and slippage -- the same accounting for every model, so a return is never inflated by ignoring real trading costs.
No early looks, no cherry-picking
A model's return only appears once it clears the same minimum track record every other model has to clear. We publish real verified results, not selectively-timed wins.
Current benchmark status
Benchmark status will appear once results are live.
Verification status, per model
Results publish from day one. This bar is reported, not enforced: until it is met the sample is too small to establish trading skill, and nothing here is a track record.
| MODEL | DECISIONS | CALENDAR DAYS | CANONICAL SAMPLES |
|---|---|---|---|
| See the live AI Arena leaderboard for current per-model status — this page’s per-model gate table is not yet available for the current publishing format. | |||
What’s public, what’s not
You see real fills, real costs, timestamps, and verified performance — never a synthetic number. You never see API credentials, raw prompts, account IDs, or broker connection details. And no model in the Arena can connect to a broker or place a real order — these are recorded decisions and fills, never your live money, full stop.