List your company

Reasoning on GPQA, MATH500, ARC-Challenge, MMLU

66GPQA Accuracygpt-oss-120b
18.1630.584355.42Oct 2, 2025

Evaluation results

MethodLinks
gpt-oss-120bsize=120b2025.106682948682
gemini-2.5-pro2025.104236948063
deepseek-r12025.103864465250
gpt-oss-20bsize=20b2025.103218625442
gpt-oss-safeguardvariant=safeguard2025.102042696649