Equity Return Recall on NASDAQ Composite monthly close-to-close
100Parse RateOpus 4.7
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Opus 4.7rho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 100 | 88 | 0.972 | 93 | |
| GPT-5.4rho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 100 | 23 | 0.71 | 78 | |
| GPT-5.4-minirho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 100 | 15 | 0.43 | 70 | |
| GPT-5.4-nanorho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 100 | 10 | 0.2 | 50 | |
| DeepSeek-V3.2rho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 100 | 23 | 0.8 | 73 | |
| Llama-3.3-70Brho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 100 | 10 | 0.18 | 60 | |
| Sonnet 4.6rho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 95 | 63 | 0.81 | 84 | |
| Haiku 4.5rho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 93 | 8 | 0.48 | 76 | |
| Llama-3.1-8Brho_FF=0.92, n (sample size)=40, Answer Format=Variant-A2026.05 | 55 | 0 | 0.03 | 50 |