Benchmarks snapshot — 2026-07-22
frozenImmutable capture of both official leaderboards and the run configuration, written once for this date and never rewritten. Generated at 2026-07-29T12:51:07.089695+00:00. 7 models · 5 instruments.
Official leaderboard — per-forecast
| # | Model | OK rate % | OK / total | Closed | Wins | PnL net $ | Equity $ | Max DD % |
|---|
| 1 | gpt-5.6-sol | 98.5 | 653 / 663 | 344 | 108 | -56.69 | 806.62 | 5.4 |
| 2 | claude-opus-4.8 | 98.5 | 653 / 663 | 428 | 135 | -60.45 | 782.17 | 6.0 |
| 3 | claude-fable-5 | 98.5 | 653 / 663 | 331 | 105 | -55.30 | 823.16 | 5.3 |
| 4 | deepseek-v4-pro | 98.5 | 653 / 663 | 266 | 84 | -43.66 | 873.95 | 4.3 |
| 5 | gemini-3.1-pro | 94.4 | 626 / 663 | 336 | 120 | -38.59 | 847.54 | 2.8 |
| 6 | qwen-3.7-max | 98.0 | 650 / 663 | 441 | 147 | -57.09 | 774.88 | 5.6 |
| 7 | grok-4.5 | 38.6 | 256 / 663 | 151 | 40 | -34.47 | 807.20 | 3.7 |
Official leaderboard — position
| # | Model | OK rate % | OK / total | Closed | Wins | PnL net $ | Equity $ | Max DD % |
|---|
| 1 | gpt-5.6-sol | 98.5 | 653 / 663 | 281 | 107 | -36.78 | 856.27 | 3.7 |
| 2 | claude-opus-4.8 | 98.5 | 653 / 663 | 322 | 114 | -37.43 | 850.06 | 3.7 |
| 3 | claude-fable-5 | 98.5 | 653 / 663 | 266 | 93 | -41.25 | 857.56 | 4.1 |
| 4 | deepseek-v4-pro | 98.5 | 653 / 663 | 229 | 77 | -32.10 | 898.02 | 3.2 |
| 5 | gemini-3.1-pro | 94.4 | 626 / 663 | 250 | 101 | -15.99 | 883.23 | 1.6 |
| 6 | qwen-3.7-max | 98.0 | 650 / 663 | 349 | 133 | -38.35 | 821.38 | 3.8 |
| 7 | grok-4.5 | 38.6 | 256 / 663 | 124 | 39 | -31.56 | 846.05 | 3.2 |
Cite this snapshot
@misc{marketmania_llmbench_s1_2026,
title = {LLM Market Benchmarks, Season 1},
author = {MarketMania},
year = {2026},
note = {Snapshot 2026-07-22, Binance USDT-M futures paper-trading},
url = {https://marketmania.ai/benchmarks/snapshots/2026-07-22}
}