Macro-average Reasoning on Overall NaturalPlan AIME 2024 GPQA
96.5Final Score (Macro-Avg)PDP
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| PDPModel=Grok-3-mini2026.01 | 96.5 | 86.9 | 88.7 | |
| LEAPModel=Grok-3-mini2026.01 | 90.9 | 84.2 | 85.9 | |
| PDP(+)Model=Grok-3-mini2026.01 | 88.6 | 83.4 | 85.1 | |
| BaseModel=Grok-3-mini2026.01 | 87.1 | 80.8 | 85.4 | |
| PDPModel=GPT-oss-120B2026.01 | 62.6 | 61.5 | 65.8 | |
| PDP(+)Model=Grok-32026.01 | 61.7 | 50.7 | 51.4 | |
| PDPModel=Grok-32026.01 | 61.6 | 51 | 51.6 | |
| LEAPModel=Grok-32026.01 | 60 | 50.7 | 51.5 | |
| BaseModel=Grok-32026.01 | 56.2 | 47.1 | 47.6 | |
| PDP(+)Model=GPT-oss-120B2026.01 | 53.7 | 56.7 | 62.5 | |
| PDPModel=GPT-4.12026.01 | 51.1 | 46.8 | 47.9 | |
| LEAPModel=Llama-3.3-70B2026.01 | 50.9 | 47 | 52.2 | |
| PDPModel=Llama-3.3-70B2026.01 | 50.8 | 50.5 | 53.9 | |
| PDP(+)Model=Llama-3.3-70B2026.01 | 50.2 | 49.5 | 53.3 | |
| PDPModel=GPT-oss-20B2026.01 | 44.7 | 44.3 | 50.7 | |
| PDPModel=GPT-4o2026.01 | 44.4 | 38.9 | 39.8 | |
| LEAPModel=GPT-4.12026.01 | 44.2 | 45.5 | 48.5 | |
| PDP(+)Model=GPT-4.12026.01 | 43.6 | 43.8 | 46.9 | |
| BaseModel=Llama-3.3-70B2026.01 | 42.6 | 47.9 | 51.8 | |
| PDP(+)Model=GPT-4o2026.01 | 42.5 | 38.7 | 39.9 | |
| BaseModel=GPT-4.12026.01 | 41 | 43.4 | 46.2 | |
| BaseModel=GPT-oss-120B2026.01 | 37.7 | 44.4 | 54.7 | |
| LEAPModel=GPT-oss-120B2026.01 | 37.7 | 42.2 | 52.9 | |
| BaseModel=GPT-4o2026.01 | 37.4 | 37 | 38.7 | |
| PDP(+)Model=GPT-oss-20B2026.01 | 35.6 | 40.5 | 48 | |
| LEAPModel=GPT-4o2026.01 | 34.7 | 34.7 | 36.8 | |
| BaseModel=GPT-oss-20B2026.01 | 32.1 | 37.5 | 48.8 | |
| LEAPModel=GPT-oss-20B2026.01 | 22.3 | 32 | 42.7 |