Recomputed from Hyperliquid's own data · seasons finished
GPT-5.1 in Alpha Arena: trading results, recomputed
In Season 1.5 its four accounts combined to −6.0%, #2 of 8 models; 1 of 4 finished up, best Max Leverage (+18.8%), worst New Baseline (−20.4%).
Every account
| Season and rule set | Return | PnL | Fees | Fills | Realized drawdown | Sources |
|---|---|---|---|---|---|---|
| Season 1.5 · Max Leverage | +18.8% | $1,876 | $1,064 | 4,232 | −24.7% | wallet · Nof1 |
| Season 1.5 · Monk Mode | −6.6% | −$659 | $279 | 2,166 | −10.7% | wallet · Nof1 |
| Season 1.5 · Situational Awareness | −15.6% | −$1,559 | $1,143 | 7,895 | −21.2% | wallet · Nof1 |
| Season 1.5 · New Baseline | −20.4% | −$2,052 | $647 | 5,435 | −22.6% | wallet · Nof1 |
FAQ
How did GPT-5.1 do in Alpha Arena?
In Season 1.5 its four accounts combined to −6.0%, #2 of 8 models; 1 of 4 finished up, best Max Leverage (+18.8%), worst New Baseline (−20.4%). Every number is recomputed from the model's Hyperliquid wallet: fills, fees, funding and positions open at the end.
Is GPT-5.1 good at trading?
Two or three weeks of trading is too short to separate skill from luck, and the same model can finish near the top in one season and near the bottom in the next. Fees mattered as much as direction: across both seasons they were 3.7 times the combined profit of the 7 accounts that finished up.