List your company
ResearchFormerly WizWand

Reasoning on AIME 2025 (Pass@1)

67.66Pass@1FP16
23.782435.173746.56557.9563May 9, 2026
Updated 5mo ago

Evaluation results

MethodLinks
FP16Model=Qwen3-8B, Speedup=1.0×2026.0567.66
LAQuantModel=Qwen3-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.45×2026.0566.98
GPTQModel=Qwen3-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.45×2026.0565.21
ParoQModel=Qwen3-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.32×2026.0564.52
ParoQ++Model=Qwen3-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.32×2026.0564.27
FP16Model=Qwen3-4B, Speedup=1.0×2026.0563.75
LAQuantModel=Qwen3-4B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.0×2026.0562.66
ParoQ++Model=Qwen3-4B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.84×2026.0561.72
ParoQModel=Qwen3-4B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.84×2026.0560.78
R-QATModel=Qwen3-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.45×2026.0559.48
GPTQModel=Qwen3-4B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.0×2026.0558.67
R-QATModel=Qwen3-4B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.0×2026.0557.66
FP16Model=Qwen3-1.7B, Speedup=1.0×2026.0535.94
LAQuantModel=Qwen3-1.7B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.64×2026.0535.47
ParoQ++Model=Qwen3-1.7B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.48×2026.0532.71
ParoQModel=Qwen3-1.7B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.48×2026.0532.55
LAQuantModel=R1-Distill-Llama-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.46×2026.0531.77
FP16Model=R1-Distill-Llama-8B, Speedup=1.0×2026.0531.2
GPTQModel=R1-Distill-Llama-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.46×2026.0529.85
R-QATModel=R1-Distill-Llama-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.46×2026.0529.53
ParoQModel=R1-Distill-Llama-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.33×2026.0528.54
ParoQ++Model=R1-Distill-Llama-8B, Quantization Bit-width=4-bit, Group Size=128, Speedup=2.33×2026.0527.6
R-QATModel=Qwen3-1.7B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.64×2026.0527.29
GPTQModel=Qwen3-1.7B, Quantization Bit-width=4-bit, Group Size=128, Speedup=1.64×2026.0525.47