Find edge AI hardware
Task medical segmentation · model-size class mid-size CNN · batch unknown · 4 platforms
Not like-for-like across groups: same model-size class and batch, but 2 different model/precision combinations. Compare within a group; each group is sorted by result.
MLPerf 3d-unet-99 · int8 (1)
| Platform | Result | Model · precision · stack | Grade · date | Power | Price | Source |
|---|---|---|---|---|---|---|
| Lenovo ThinkEdge SE455i V3 (2x NVIDIA L4-PCIe-24GB) | 2.19 FPS throughput_offline b=unknown | MLPerf 3d-unet-99 int8 · TensorRT 10.8, CUDA 13.1 | A 2026-04-01 (published) | unknown no power published | unknown | MLPerf Inference v6.0, closed division, edge suite |
MLPerf 3D-Unet-99.0 · INT8/FP16 (MLPerf closed) (3)
| Platform | Result | Model · precision · stack | Grade · date | Power | Price | Source |
|---|---|---|---|---|---|---|
| NVIDIA Jetson AGX Orin 64GB | 0.510 FPS throughput_offline b=unknown | MLPerf 3D-Unet-99.0 INT8/FP16 (MLPerf closed) · TensorRT (MLPerf Inference, older round) | D 2023-09-08 (published) | 65.00 W platform spec envelope (not measured with this result) | $3499 | NVIDIA Jetson Benchmarks (MLPerf v3.1 results table) |
| NVIDIA Jetson Orin NX 16GB | 0.200 FPS throughput_offline b=unknown | MLPerf 3D-Unet-99.0 INT8/FP16 (MLPerf closed) · TensorRT (MLPerf Inference, older round) | D 2023-09-08 (published) | unknown no power published | unknown | NVIDIA Jetson Benchmarks (MLPerf v3.1 results table) |
| NVIDIA Jetson Orin NX 16GB (MaxQ) | 0.190 FPS throughput_offline b=unknown | MLPerf 3D-Unet-99.0 INT8/FP16 (MLPerf closed) · TensorRT (MLPerf Inference, older round) | D 2023-09-08 (published) | 22.04 W reported in result source — measured | unknown | NVIDIA Jetson Benchmarks (MLPerf v3.1 results 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)