Equity Return Recall on S&P 500 monthly close-to-close
1Perfect ScoreOpus 4.7
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Opus 4.7rho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 100 | 1 | 100 | |
| Sonnet 4.6rho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 85 | 0.97 | 95 | |
| Haiku 4.5rho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 38 | 0.59 | 75 | |
| GPT-5.4rho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 63 | 0.91 | 88 | |
| GPT-5.4-minirho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 50 | 0.76 | 83 | |
| GPT-5.4-nanorho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 8 | 0.43 | 60 | |
| DeepSeek-V3.2rho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 55 | 0.86 | 83 | |
| Llama-3.3-70Brho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 45 | 0.68 | 65 | |
| Llama-3.1-8Brho_FF=0.99, n (sample size)=40, Answer Format=Variant-A2026.05 | 1 | 3 | 0.23 | 40 |