List your company

Decoding Speedup on InfiniteBench (Long-context understanding)

1.48Decoding SpeedupPRR
1.26161.31831.3751.4317Jun 29, 2026

Evaluation results

MethodLinks
PRRModel=DeepSeek-R1, DSA mechanism=InfLLM-v22026.061.48
PRRModel=Llama3, DSA mechanism=InfLLM-v22026.061.45
PRRModel=Qwen3-14B, DSA mechanism=InfLLM-v22026.061.43
PRRModel=DeepSeek-R1, DSA mechanism=Quest2026.061.42
PRRModel=GLM-4, DSA mechanism=Quest2026.061.4
PRRModel=Llama3, DSA mechanism=Quest2026.061.4
PRRModel=GLM-Z1, DSA mechanism=Quest2026.061.39
PRRModel=GLM-4, DSA mechanism=InfLLM-v22026.061.38
PRRModel=GLM-Z1, DSA mechanism=InfLLM-v22026.061.35
PRRModel=Qwen3-32B, DSA mechanism=Quest2026.061.34
PRRModel=Qwen3-14B, DSA mechanism=Quest2026.061.27
PRRModel=Qwen3-32B, DSA mechanism=InfLLM-v22026.061.27