List your company

Massively Multitask Language Understanding on MMLU-pro (Performance, Rank)

43PerformanceGPT-5
10.7619.1327.535.87Jan 9, 2026

Evaluation results

MethodLinks
GPT-5Temperature=0, Access=Official API2026.01431
Qwen2.5-72BTemperature=0, Access=Official API2026.01272
Qwen2.5-32BTemperature=0, Access=Official API2026.01233
Qwen-turboTemperature=0, Access=Official API2026.01154
Llama3-8BTemperature=0, Hardware=2 4090 GPUs with 24GB memory2026.01125