LLM Inference on MBPP
3.09SpeedupDART
Evaluation Results
| Method | Links | |
|---|---|---|
| DARTModel=Qwen8B, Temperature=02026.01 | 3.09 | |
| DARTModel=Qwen4B, Temperature=02026.01 | 3.06 | |
| DARTModel=Qwen14B, Temperature=02026.01 | 2.98 | |
| DARTModel=L2 7B, Temperature=02026.01 | 2.72 | |
| DARTModel=Qwen14B, Temperature=12026.01 | 2.71 | |
| DARTModel=Qwen1.7B, Temperature=02026.01 | 2.64 | |
| HydraModel=L2 7B, Temperature=02026.01 | 2.62 | |
| DARTModel=Qwen32B, Temperature=02026.01 | 2.56 | |
| DARTModel=Qwen32B, Temperature=12026.01 | 2.4 | |
| MedusaModel=L2 7B, Temperature=02026.01 | 2.35 | |
| EAGLE3Model=Qwen32B, Temperature=02026.01 | 2.27 | |
| EAGLE3Model=Qwen8B, Temperature=02026.01 | 2.2 | |
| EAGLE3Model=Qwen32B, Temperature=12026.01 | 2.17 | |
| EAGLE3Model=Qwen4B, Temperature=02026.01 | 2.08 | |
| EAGLE3Model=Qwen1.7B, Temperature=02026.01 | 2.05 | |
| EAGLE3Model=Qwen14B, Temperature=02026.01 | 1.99 | |
| EAGLE3Model=Qwen14B, Temperature=12026.01 | 1.81 | |
| LookaheadModel=L2 7B, Temperature=02026.01 | 1.61 | |
| PLDModel=L2 7B, Temperature=02026.01 | 1.59 | |
| SPSModel=Qwen32B, Temperature=02026.01 | 1.15 | |
| SPSModel=Qwen14B, Temperature=02026.01 | 0.97 |