Find edge AI hardware
Task super-resolution · model-size class mid-size CNN · batch unknown · 4 platforms
✓ Like-for-like: same model (pt_DRUNet_Kvasir_528_608_0.4G), precision (INT8), model-size class and batch.
| Platform | Result | Model · precision · stack | Grade · date | Power | Price | Source |
|---|---|---|---|---|---|---|
| AMD Versal AI Core VCK190 (VC1902) | 462 FPS E2E throughput, multi thread b=unknown | pt_DRUNet_Kvasir_528_608_0.4G INT8 · Vitis AI 3.0 (VART) | D 2023-01-13 (published) | unknown no power published | unknown | Vitis AI 3.0 Model Zoo performance table |
| AMD Zynq UltraScale+ ZCU102 (ZU9EG) | 191 FPS E2E throughput, multi thread b=unknown | pt_DRUNet_Kvasir_528_608_0.4G INT8 · Vitis AI 3.0 (VART) | D 2023-01-13 (published) | unknown no power published | unknown | Vitis AI 3.0 Model Zoo performance table |
| AMD Zynq UltraScale+ ZCU104 (ZU7EV) | 152 FPS E2E throughput, multi thread b=unknown | pt_DRUNet_Kvasir_528_608_0.4G INT8 · Vitis AI 3.0 (VART) | D 2023-01-13 (published) | unknown no power published | unknown | Vitis AI 3.0 Model Zoo performance table |
| AMD Kria KV260 Vision AI Starter Kit (K26 SOM) | 77.69 FPS E2E throughput, multi thread b=unknown | pt_DRUNet_Kvasir_528_608_0.4G INT8 · Vitis AI 3.0 (VART) | D 2023-01-13 (published) | 7.50 W reported in result source — spec | unknown | Vitis AI 3.0 Model Zoo performance table |
Confidence: A audited MLPerf closed, B documented third-party, C community, D vendor-published, E estimate. Unknown is never assumed. Other like-for-like groups: mid-size CNN, batch not stated (4); unknown, batch 1 (2); unknown, batch not stated (6)