instance segmentation benchmarks on edge AI hardware
Each table compares only results with the same model-size class and batch size. Search this workload · JSON.
Fastest in the best-covered like-for-like group (nano (YOLO-n), batch 1): Hailo-8 (M.2/PCIe, x86 host) at 528 FPS D undated. 7 groups, 9 platforms.
Show data — 7 like-for-like tables
nano (YOLO-n) · batch 1 (5 platforms)
✓ Like-for-like: model-size class nano (YOLO-n), batch 1. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Hailo-8 (M.2/PCIe, x86 host) | 528 FPS throughput · b=1 | yolov8n_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · nano (YOLO-n) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8) undated |
| Hailo-10H (M.2/PCIe, x86 host) | 311 FPS throughput · b=1 | yolov8n_seg INT8 (Hailo quantized) · Hailo DFC v5.4.0 / HailoRT (model zoo) · nano (YOLO-n) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-10H) undated |
| Apple M4 (10-core GPU, 120 GB/s) | 196 FPS latency_e2e · b=1 | YOLO26n-seg FP16 · Ultralytics 8.4.168 Core AI (coreai-torch 0.4.3), macOS 27.0 · nano (YOLO-n) | B | unknown power n/a no power published | Ultralytics docs: Apple Core AI integration benchmark 2026 (year only) |
| Hailo-8L (M.2/PCIe, x86 host) | 168 FPS throughput · b=1 | yolov8n_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · nano (YOLO-n) | D | 1.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8L) undated |
| Microchip PolarFire SoC Video Kit (VectorBlox CoreVectorBlox) | 14.13 FPS inference_time · b=1 | yolov8n-seg INT8 (VNNX) · VectorBlox SDK (VNNX), no compression · nano (YOLO-n) | D | unknown power n/a no power published | Microchip VectorBlox SDK tutorial metrics undated |
medium (YOLO-m) · batch 1 (3 platforms)
✓ Like-for-like: model-size class medium (YOLO-m), batch 1. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Hailo-8 (M.2/PCIe, x86 host) | 68.70 FPS throughput · b=1 | yolov5m_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · medium (YOLO-m) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8) undated |
| Hailo-10H (M.2/PCIe, x86 host) | 66.70 FPS throughput · b=1 | yolov8m_seg INT8 (Hailo quantized) · Hailo DFC v5.4.0 / HailoRT (model zoo) · medium (YOLO-m) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-10H) undated |
| Hailo-8L (M.2/PCIe, x86 host) | 52.90 FPS throughput · b=1 | yolov5m_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · medium (YOLO-m) | D | 1.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8L) undated |
medium (YOLO-m) · batch 8 (3 platforms)
✓ Like-for-like: model-size class medium (YOLO-m), batch 8. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Hailo-8 (M.2/PCIe, x86 host) | 125 FPS throughput · b=8 | yolov5m_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · medium (YOLO-m) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8) undated |
| Hailo-10H (M.2/PCIe, x86 host) | 94.50 FPS throughput · b=8 | yolov8m_seg INT8 (Hailo quantized) · Hailo DFC v5.4.0 / HailoRT (model zoo) · medium (YOLO-m) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-10H) undated |
| Hailo-8L (M.2/PCIe, x86 host) | 83.60 FPS throughput · b=8 | yolov5m_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · medium (YOLO-m) | D | 1.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8L) undated |
nano (YOLO-n) · batch 8 (3 platforms)
✓ Like-for-like: model-size class nano (YOLO-n), batch 8. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Hailo-8 (M.2/PCIe, x86 host) | 528 FPS throughput · b=8 | yolov8n_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · nano (YOLO-n) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8) undated |
| Hailo-8L (M.2/PCIe, x86 host) | 338 FPS throughput · b=8 | yolov8n_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · nano (YOLO-n) | D | 1.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8L) undated |
| Hailo-10H (M.2/PCIe, x86 host) | 212 FPS throughput · b=8 | yolov8n_seg INT8 (Hailo quantized) · Hailo DFC v5.4.0 / HailoRT (model zoo) · nano (YOLO-n) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-10H) undated |
small (YOLO-s) · batch 1 (3 platforms)
✓ Like-for-like: model-size class small (YOLO-s), batch 1. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Hailo-8 (M.2/PCIe, x86 host) | 331 FPS throughput · b=1 | yolov5s_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · small (YOLO-s) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8) undated |
| Hailo-10H (M.2/PCIe, x86 host) | 177 FPS throughput · b=1 | yolov5s_seg INT8 (Hailo quantized) · Hailo DFC v5.4.0 / HailoRT (model zoo) · small (YOLO-s) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-10H) undated |
| Hailo-8L (M.2/PCIe, x86 host) | 101 FPS throughput · b=1 | yolov5s_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · small (YOLO-s) | D | 1.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8L) undated |
small (YOLO-s) · batch 8 (3 platforms)
✓ Like-for-like: model-size class small (YOLO-s), batch 8. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Hailo-8 (M.2/PCIe, x86 host) | 331 FPS throughput · b=8 | yolov5s_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · small (YOLO-s) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8) undated |
| Hailo-10H (M.2/PCIe, x86 host) | 162 FPS throughput · b=8 | yolov8s_seg INT8 (Hailo quantized) · Hailo DFC v5.4.0 / HailoRT (model zoo) · small (YOLO-s) | D | 2.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-10H) undated |
| Hailo-8L (M.2/PCIe, x86 host) | 160 FPS throughput · b=8 | yolov5s_seg INT8 (Hailo quantized) · Hailo DFC v2.20.0 / HailoRT (model zoo) · small (YOLO-s) | D | 1.50 W spec reported in result source — spec | Hailo Model Zoo public models (Hailo-8L) undated |
unknown · batch 1 (4 platforms)
✓ Like-for-like: model-size class unknown, batch 1 (size or batch not stated by source — unverified). Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Qualcomm Snapdragon X2 Elite (CRD) | 1311 FPS inference_time · b=1 | yolov8_seg / YOLOv8-Segmentation w8a8 · Qualcomm AI Hub onnx (QAIRT 2.50.0.260828221209) · unknown | D | unknown power n/a no power published | Qualcomm AI Hub Models perf.yaml (BSD-3) 2026 (year only) |
| Qualcomm Dragonwing IQ-9075 EVK | 568 FPS inference_time · b=1 | yolov8_seg / YOLOv8-Segmentation w8a8 · Qualcomm AI Hub tflite (QAIRT 2.50.0.260828221209) · unknown | D | unknown power n/a no power published | Qualcomm AI Hub Models perf.yaml (BSD-3) 2026 (year only) |
| Qualcomm Snapdragon X Elite (CRD) | 556 FPS inference_time · b=1 | yolov8_seg / YOLOv8-Segmentation w8a8 · Qualcomm AI Hub qnn_dlc (QAIRT 2.50.0.260828221209) · unknown | D | unknown power n/a no power published | Qualcomm AI Hub Models perf.yaml (BSD-3) 2026 (year only) |
| Qualcomm Dragonwing RB3 Gen 2 (QCS6490) | 240 FPS inference_time · b=1 | yolov8_seg / YOLOv8-Segmentation w8a8 · Qualcomm AI Hub tflite (QAIRT 2.50.0.260828221209) · unknown | D | unknown power n/a no power published | Qualcomm AI Hub Models perf.yaml (BSD-3) 2026 (year only) |
How to read these numbers
Every number is labelled. Confidence: A MLPerf closed, audited · B documented third-party method · C community / review · D vendor-published, not independently verified · E estimate. Power basis: measured with the result · spec vendor spec/envelope · power n/a not published.
Machine-readable: /find.json (see llms.txt); full tables are in “Show data”. Confidence grades: A audited MLPerf closed · B documented third-party · C community · D vendor-published · E estimate. Unknown values are shown as unknown, never estimated. Generated 2026-10-16 from public sources.