Decoding Speedup on LongBench
1.64Decoding SpeedupPRR
Evaluation Results
| Method | Links | |
|---|---|---|
| PRRModel=GLM-4, DSA mechanism=InfLLM-v22026.06 | 1.64 | |
| PRRModel=GLM-Z1, DSA mechanism=InfLLM-v22026.06 | 1.61 | |
| PRRModel=Qwen3-14B, DSA mechanism=InfLLM-v22026.06 | 1.56 | |
| PRRModel=GLM-4, DSA mechanism=Quest2026.06 | 1.5 | |
| PRRModel=GLM-Z1, DSA mechanism=Quest2026.06 | 1.48 | |
| PRRModel=Qwen3-14B, DSA mechanism=Quest2026.06 | 1.44 | |
| PRRModel=DeepSeek-R1, DSA mechanism=Quest2026.06 | 1.39 | |
| PRRModel=Llama3, DSA mechanism=Quest2026.06 | 1.36 | |
| PRRModel=Qwen3-32B, DSA mechanism=InfLLM-v22026.06 | 1.35 | |
| PRRModel=Llama3, DSA mechanism=InfLLM-v22026.06 | 1.32 | |
| PRRModel=DeepSeek-R1, DSA mechanism=InfLLM-v22026.06 | 1.3 | |
| PRRModel=Qwen3-32B, DSA mechanism=Quest2026.06 | 1.26 |