Hybrid Mamba-Transformer Inference on workload uniform_short
434.24Goodput (req/s)avmp_dynamic_b128
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| avmp_dynamic_b128variant=avmp_dynamic_b1282026.05 | 434.24 | 13.3 | |
| fixed_dual_mr05variant=fixed_dual_mr052026.05 | 32.65 | — |
| Method | Links | ||
|---|---|---|---|
| avmp_dynamic_b128variant=avmp_dynamic_b1282026.05 | 434.24 | 13.3 | |
| fixed_dual_mr05variant=fixed_dual_mr052026.05 | 32.65 | — |