Large Language Model Inference on Decode Phase BS=1
0.152Latency (s)
BWTA (Bitnet-b1.58-2B)
Updated 11d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.04 | 0.152 | 375.4 | |
2026.04 | 0.156 | 365.9 | |
2026.04 | 0.158 | 360.2 | |
2026.04 | 0.285 | 344.4 | |
2026.04 | 0.29 | 338 | |
2026.04 | 0.295 | 331 | |
2026.04 | 0.443 | 345.8 | |
2026.04 | 0.449 | 338.4 | |
2026.04 | 0.458 | 333.9 | |
2026.04 | 0.512 | 111.6 | |
2026.04 | 0.989 | 50.8 | |
2026.04 | 1.146 | 87.3 | |
2026.04 | 1.317 | 43.4 | |
2026.04 | 1.86 | 83.3 | |
2026.04 | 1.959 | 51 | |
2026.04 | 2.563 | 39 | |
2026.04 | 2.932 | 52.9 | |
2026.04 | 3.507 | 44.2 |