List your company

Knowledge-intensive Reasoning on SuperGPQA

72.7Overall ScoreGemini-3-Flash
16.43631.04345.6560.257Oct 9, 2025Nov 22, 2025Jan 5, 2026Feb 18, 2026Apr 3, 2026May 17, 2026Jun 30, 2026

Evaluation results

MethodLinks
Gemini-3-FlashTier=High2026.0672.7
Seed2.0 Lite2026.0667.5
Qwen3.5Parameter Count=35BA3B2026.0563.4
Gemma4Parameter Count=26BA4B2026.0561.88
Seed2.0 Mini2026.0661.6
GPT-5-miniTier=High2026.0660.5
SenseNova-U1Parameter Count=30BA3B, Configuration Variant=Think2026.0559.71
Qwen3.5Parameter Count=9B2026.0558.2
Qwen3VLParameter Count=30BA3B, Configuration Variant=Think2026.0556.4
Qwen3VLParameter Count=8B, Configuration Variant=Think2026.0551.2
SenseNova-U1Parameter Count=8B, Configuration Variant=Think2026.0549.67
SCOPE-RLBase Model=Qwen3-4B2025.1038.6
GRPOBase Model=Qwen3-4B2025.1038.4
Entropy-RegBase Model=Qwen3-4B2025.1038.2
Entropy-AdvBase Model=Qwen3-4B2025.1037.8
KL-covBase Model=Qwen3-4B2025.1037.6
DAPOBase Model=Qwen3-4B2025.1037
CISPOBase Model=Qwen3-4B2025.1036.8
BaseBase Model=Qwen3-4B2025.1032.8
SCOPE-RLBase Model=Qwen2.5-7B2025.1027.2
GRPOBase Model=Qwen2.5-7B2025.1027
SCOPE-RLBase Model=Qwen2.5-Math-7B2025.1026.8
Entropy-RegBase Model=Qwen2.5-7B2025.1026.8
KL-covBase Model=Qwen2.5-Math-7B2025.1026.6
Entropy-AdvBase Model=Qwen2.5-7B2025.1026.6
Entropy-RegBase Model=Qwen2.5-Math-7B2025.1026.2
Entropy-AdvBase Model=Qwen2.5-Math-7B2025.1026.2
CISPOBase Model=Qwen2.5-Math-7B2025.1026.2
DAPOBase Model=Qwen2.5-Math-7B2025.1026
KL-covBase Model=Qwen2.5-7B2025.1026
GRPOBase Model=Qwen2.5-Math-7B2025.1025.8
CISPOBase Model=Qwen2.5-7B2025.1025.8
DAPOBase Model=Qwen2.5-7B2025.1024.8
BaseBase Model=Qwen2.5-Math-7B2025.1019.4
BaseBase Model=Qwen2.5-7B2025.1018.6