Decoding Speedup on RULER
1.61Decoding SpeedupPRR
Evaluation Results
| Method | Links | |
|---|---|---|
| PRRModel=GLM-Z1, DSA mechanism=InfLLM-v22026.06 | 1.61 | |
| PRRModel=DeepSeek-R1, DSA mechanism=InfLLM-v22026.06 | 1.61 | |
| PRRModel=GLM-4, DSA mechanism=InfLLM-v22026.06 | 1.58 | |
| PRRModel=Llama3, DSA mechanism=InfLLM-v22026.06 | 1.55 | |
| PRRModel=Qwen3-14B, DSA mechanism=InfLLM-v22026.06 | 1.51 | |
| PRRModel=DeepSeek-R1, DSA mechanism=Quest2026.06 | 1.47 | |
| PRRModel=GLM-4, DSA mechanism=Quest2026.06 | 1.45 | |
| PRRModel=GLM-Z1, DSA mechanism=Quest2026.06 | 1.43 | |
| PRRModel=Llama3, DSA mechanism=Quest2026.06 | 1.41 | |
| PRRModel=Qwen3-14B, DSA mechanism=Quest2026.06 | 1.41 | |
| PRRModel=Qwen3-32B, DSA mechanism=InfLLM-v22026.06 | 1.29 | |
| PRRModel=Qwen3-32B, DSA mechanism=Quest2026.06 | 1.27 |