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

Mathematics on MATH-500 (test)

80.2Score (%)

Full

Updated 1mo ago

Evaluation Results

Method	Links
Full 2026.06		80.2
MSAN 2026.06		79.6
MAN 2026.06		79.2
Full 2026.06		78
REAP 2026.06		77.8
MAN 2026.06		58
MSAN 2026.06		56.4
FReDA-4B 2026.06		53
REAP 2026.06		52.8
Qwen3-4B-Base 2026.06		52.6
FReDA-4B 2026.06		51.98
BlockDiff-4B 2026.06		50.6
Qwen2.5-3B-Base 2026.06		42.6
LLaDA-MoE-7B-A1B-Base 2026.06		41.2
LLaDA-8B-Base 2026.06		39.6
Dream-7B-Base 2026.06		39.2
MoNE 2026.06		38.4
EAN 2026.06		37
SEER 2026.06		36.8
Frequency 2026.06		31.6
Full 2026.06		29.8
Full 2026.06		22.2
MoNE 2026.06		20.6
SEER 2026.06		20.2
Frequency 2026.06		19.6
EAN 2026.06		19.4
DiffuLLaMA-7B 2026.06		11
MSAN 2026.06		10.2
MSAN 2026.06		5.6
MAN 2026.06		4.6
REAP 2026.06		3.6
MAN 2026.06		3.2
EAN 2026.06		2.8
REAP 2026.06		2.8
Frequency 2026.06		2.4
MoNE 2026.06		2.4
SEER 2026.06		1.6
SEER 2026.06		1.2
EAN 2026.06		1.2
Frequency 2026.06		1.2
MoNE 2026.06		0.6