List your company

Aggregate Benchmark Performance on Math500, GSM8K, MMLU-s, GPQA, AIME Suite

42.99Average P@1ICT
4.156414.238224.3234.4018Jun 18, 2026

Evaluation results

MethodLinks
ICTModel Size=Qwen2.5-7B2026.0642.9958.3750.68
STAPOModel Size=Qwen2.5-7B2026.0641.5555.5248.54
GRPOModel Size=Qwen2.5-7B2026.0639.453.1646.28
BaseModel Size=Qwen2.5-7B2026.0637.7855.2546.52
20-EntropyModel Size=Qwen2.5-7B2026.0637.5353.2945.41
ICTModel Size=Qwen2.5-1.5B2026.0621.9336.5329.23
STAPOModel Size=Qwen2.5-1.5B2026.0619.8931.5225.71
20-EntropyModel Size=Qwen2.5-1.5B2026.0618.6631.0224.84
GRPOModel Size=Qwen2.5-1.5B2026.0618.2231.8725.05
BaseModel Size=Qwen2.5-1.5B2026.0616.4131.824.11
ICTModel Size=Qwen2.5-0.5B2026.0610.9521.7516.35
GRPOModel Size=Qwen2.5-0.5B2026.0610.5218.9714.75
STAPOModel Size=Qwen2.5-0.5B2026.069.5418.2913.92
20-EntropyModel Size=Qwen2.5-0.5B2026.068.2616.3712.32
BaseModel Size=Qwen2.5-0.5B2026.065.6514.4810.07