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

Mathematical Reasoning on AIME 2025 (Avg@1/8 e_m, Pass@1/8 e_m)

72.9Avg@8 (e_m)

GPT-5 nano

Updated 20d ago

Evaluation Results

Method	Links
GPT-5 nano 2026.05		72.9	73.3	73.3	86.7
gpt-oss-20b 2026.05		66.7	76.7	76.7	93.3
Qwen3-8B 2026.05		61.2	63.3	63.3	76.7
Qwen3-4B 2026.05		58.3	60	60	70
EngGPT2-16B-A3B 2026.05		31.7	30	30	50
Ministral-3-8B 2026.05		26.7	26.7	26.7	26.7
GRPO 2026.05		25.8	-	-	36.66
MOPD 2026.05		25.41	-	-	36.28
Qwen3-4B 2026.05		17.92	-	-	32.57
gemma-3-12b-it 2026.05		17.5	13.3	13.3	20
gemma-3-4b-it 2026.05		13.3	13.3	13.3	13.3
Moonlight-16B-A3B-Instruct 2026.05		9.6	6.7	6.7	10
SDPO 2026.05		7.81	-	-	16.66
FastwebMIIA-7B 2026.05		0	0	0	0
Minerva-7B-instruct-v1.0 2026.05		0	0	0	0
Velvet-14B 2026.05		0	0	0	0
LLaMAntino-3-ANITA-8B 2026.05		0	0	0	0
Llama-3.2-3B-Instruct 2026.05		0	0	0	0
Llama-3.1-8B-Instruct 2026.05		0	0	0	0
deepseek-moe-16b-chat 2026.05		0	0	0	0