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

Multilingual Math Reasoning on MGSM (Mean@3)

84.43Mean@3

x1-Qwen3-32B

Updated 1mo ago

Evaluation Results

Method	Links
x1-Qwen3-32B 2026.04		84.43
Qwen3-32B 2026.04		83.98
x1-Qwen3-14B 2026.04		83.64
Qwen3-14B 2026.04		82.56
o4-mini-high 2026.04		82.32
Qwen3-32B 2026.04		80.52
x1-Qwen3-32B 2026.04		80.12
DeepSeek-V3.2 2026.04		79.24
x1-Qwen3-4B 2026.04		77.69
Qwen3-14B 2026.04		77.64
x1-Qwen3-14B 2026.04		77.38
Qwen3-4B 2026.04		76.59
DeepSeek-V3.2 2026.04		76.32
x1-Qwen3-4B 2026.04		70.3
Qwen3-4B 2026.04		70.21
x1-DeepSeek-R1-Distill-Qwen-7B 2026.04		63.24
DeepSeek-R1-Distill-Qwen-7B 2026.04		60.05
DeepSeek-R1-Distill-Qwen-7B 2026.04		54.76
x1-DeepSeek-R1-Distill-Qwen-7B 2026.04		54.52
x1-DeepSeek-R1-Distill-Llama-8B 2026.04		52.17
DeepSeek-R1-Distill-Llama-8B 2026.04		40.36
DeepSeek-R1-Distill-Llama-8B 2026.04		38.17
x1-DeepSeek-R1-Distill-Llama-8B 2026.04		38.01