List your company

Scientific QA on GPQA-Diamond

99.3Final AccuracyBase
17.86839.00960.1581.291Jan 16, 2026

Evaluation results

MethodLinks
BaseModel=Grok-3-mini2026.0199.390.490.6
PDPModel=Grok-3-mini2026.0198.989.289.3
PDP(+)Model=Grok-3-mini2026.0196.985.985.8
LEAPModel=Grok-3-mini2026.0195.79090.6
PDP(+)Model=Grok-32026.0170.363.763.8
BaseModel=Grok-32026.0169.863.563.5
PDPModel=Grok-32026.0169.764.364.4
PDPModel=GPT-oss-120B2026.0169.466.266.2
PDPModel=GPT-4.12026.0169.466.267.1
LEAPModel=Grok-32026.0169.363.463.4
PDPModel=GPT-4o2026.0165.354.854.9
PDPModel=GPT-oss-20B2026.0160.756.460.7
PDP(+)Model=GPT-4o2026.015952.453
LEAPModel=GPT-4o2026.0156.45253.1
PDP(+)Model=GPT-4.12026.0152.957.262.3
BaseModel=GPT-4o2026.0152.85354.7
BaseModel=GPT-4.12026.0150.957.761.7
PDP(+)Model=GPT-oss-120B2026.0150.654.462.5
PDPModel=Llama-3.3-70B2026.0148.147.952.4
LEAPModel=GPT-4.12026.0147.758.263.7
PDP(+)Model=Llama-3.3-70B2026.0145.447.352.8
BaseModel=GPT-oss-120B2026.0144.351.563.8
LEAPModel=Llama-3.3-70B2026.0142.844.752
BaseModel=Llama-3.3-70B2026.0139.346.452.5
LEAPModel=GPT-oss-120B2026.0136.847.963.6
PDP(+)Model=GPT-oss-20B2026.0134.247.358.2
BaseModel=GPT-oss-20B2026.0128.443.558.2
LEAPModel=GPT-oss-20B2026.012136.856.7