List your company

Reasoning and Math Suite on GSM8K, CommonSense, BoolQ, ARC Challenge, and HellaSwag

87.8Average AccuracySELF-REDTEAM
66.27271.86177.4583.039May 8, 2026

Evaluation results

MethodLinks
SELF-REDTEAMBackbone=Qwen2.5-14B-IT2026.0587.8
Qwen2.5-14B-ITBackbone=Qwen2.5-14B-IT2026.0587.4
ABSBackbone=Qwen2.5-14B-IT2026.0587.4
Qwen2.5-7B-ITBackbone=Qwen2.5-7B-IT2026.0585
ABSBackbone=Qwen2.5-7B-IT2026.0584.7
SELF-REDTEAMBackbone=Qwen2.5-7B-IT2026.0581.8
Qwen2.5-3B-ITBackbone=Qwen2.5-3B-IT2026.0574.5
ABSBackbone=Qwen2.5-3B-IT2026.0573.8
SELF-REDTEAMBackbone=Qwen2.5-3B-IT2026.0567.1