Strategy 1 is a simple trading strategy where we task the model with buying and selling MSFT stock at predetermined times. We take Databento's market-by-price L10 data for MSFT. We then randomize the volume available at each price level to enforce correct synthetic-book reservation. If we schedule a trade for 10 shares and only 8 exist in the book, the model must only take 8. This caps the max volume available in the book.
The strategy tracks: Cash and MSFT position; Realized P&L using FIFO accounting; Unrealized P&L based on raw-book mid-prices; An equity curve and maximum drawdown; Synthetic-book statistics such as total size available and bid/ask VWAP post model trades.
Loading chart data...
| Model | Mean MAE↓ | Best Run MAE↓ | Executable Passes↓ |
|---|---|---|---|
| gemini-3-pro-preview | 14.83 | 14.83 | 5 |
| claude-sonnet-4.5 | 16.36 | 16.35 | 5 |
| claude-opus-4.5 | 111.40 | 7.18 | 3 |
| grok-4 | 121.85 | 7.22 | 4 |
| deepseek-v3.2 | 139.19 | 7.26 | 5 |
| command-a | 184.90 | 153.10 | 1 |
| nova-premier-v1 | 242.34 | 153.10 | 2 |
| mistral-large-2512 | 361.87 | 23.01 | 5 |
| gpt-5.1-codex-max | 844.74 | 0.002 | 5 |
| gpt-5.2 | 1,545.89 | 153.10 | 5 |
| qwen3-max | 2,015.66 | 16.39 | 5 |
| llama-3.1-nemotron-ultra | 2,511.13 | 111.63 | 3 |
| llama-4-maverick | 4,137.62 | 170.06 | 5 |