QuantCodeEval (Codex CLI): leaderboard
QuantCodeEval — reproducing quantitative-finance strategies from 30 research papers, run under the Codex CLI agent harness. Full Pass Rate (%) counts an attempt only when every one of the 572 property checkers in the task contract passes, verifying each pipeline stage rather than the backtest metric alone.
Metric: Full Pass Rate (%). Source: www.quantcodeeval.cloud. Saturation forecast: Around 2028. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 27.33 | #22 |
| 2 | GPT-5.5 | 24.67 | #33 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=quantcodeeval-codex-cli · How It Works · Data refreshed daily, snapshot 2026-11-05.