Share your thoughts, 1 month free Claude Pro on usSee more

Knowledge-intensive Reasoning on SuperGPQA

72.7Overall Score

Gemini-3-Flash

Updated 23d ago

Evaluation Results

Method	Links
Gemini-3-Flash 2026.06		72.7
Seed2.0 Lite 2026.06		67.5
Qwen3.5 2026.05		63.4
Gemma4 2026.05		61.88
Seed2.0 Mini 2026.06		61.6
GPT-5-mini 2026.06		60.5
SenseNova-U1 2026.05		59.71
Qwen3.5 2026.05		58.2
Qwen3VL 2026.05		56.4
Qwen3VL 2026.05		51.2
SenseNova-U1 2026.05		49.67
SCOPE-RL 2025.10		38.6
GRPO 2025.10		38.4
Entropy-Reg 2025.10		38.2
Entropy-Adv 2025.10		37.8
KL-cov 2025.10		37.6
DAPO 2025.10		37
CISPO 2025.10		36.8
Base 2025.10		32.8
SCOPE-RL 2025.10		27.2
GRPO 2025.10		27
SCOPE-RL 2025.10		26.8
Entropy-Reg 2025.10		26.8
KL-cov 2025.10		26.6
Entropy-Adv 2025.10		26.6
Entropy-Reg 2025.10		26.2
Entropy-Adv 2025.10		26.2
CISPO 2025.10		26.2
DAPO 2025.10		26
KL-cov 2025.10		26
GRPO 2025.10		25.8
CISPO 2025.10		25.8
DAPO 2025.10		24.8
Base 2025.10		19.4
Base 2025.10		18.6