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

Mathematical Reasoning on BeyondAIME (accuracy)

82.5Accuracy

TRICE-30B

Updated 2mo ago

Evaluation Results

Method	Links
TRICE-30B 2026.05		82.5
TRICE-30B 2026.05		82.5
GLM-4.7-Flash w/ recipe 2026.05		81
Nemotron-3-Nano-30B-A3B 2026.05		80
DeepSeek-V3.2-Thinking 2026.05		76.8
GLM-4.7-Flash 2026.05		76
Qwen3.5-35B-A3B 2026.05		72.5
Qwen3-235B-A22B-Thinking 2026.05		71.8
TRICE-4B 2026.05		71.3
TRICE-30B 2026.05		71
Qwen3.5-9B 2026.05		67.3
Qwen3-30B-A3B-Thinking-2507 2026.05		65.9
GPT-OSS-20B 2026.05		63
ASTER-4B† 2026.05		61.7
Qwen3.5-4B 2026.05		58.8
TRICE-4B 2026.05		58.5
Qwen3-4B-Thinking-2507 2026.05		54.3
Qwen3-30B-A3B-Instruct-2507 2026.05		51.3