Reasoning on AIME

73.33AIME Score

DeepSeek-R1-Distill-Qwen-32B (Reasoning)

Updated 3d ago

Evaluation Results

Method	Links
DeepSeek-R1-Distill-Qwen-32B (Reasoning) 2026.01		73.33
DeepSeek-R1-Distill-Qwen-14B (Reasoning) 2026.01		63.33
ReasonAny 2026.01		56.67
ReasonAny 2026.01		46.67
Task Arithmetic 2026.01		36.67
TIES 2026.01		33.33
Reasoning 2026.01		33.33
Task Arithmetic 2026.01		30
LED 2026.01		30
Qwen2.5-32B-Instruct (Safety) 2026.01		23.33
DARE 2026.01		23.33
FuseLLM 2026.01		23.33
LED 2026.01		20
Qwen2.5-14B-Instruct (Safety) 2026.01		20
Linear 2026.01		16.67
DARE 2026.01		16.67
ReasonAny 2026.01		13.33
Linear 2026.01		13.33
FuseLLM 2026.01		10
Linear 2026.01		6.67
TIES 2026.01		6.67
TIES 2026.01		6.67
Safety 2026.01		0
Task Arithmetic 2026.01		0
DARE 2026.01		0
FuseLLM 2026.01		0
LED 2026.01		0