YOLOv8s on Hailo-8: 2 rows.
Method v1.0 · dataset 2026-09-07 · verified 2026-09-07 · last updated September 2026
yolov8s on Hailo-8: 2 rows (1 measured, 1 derived). Expected: INT8 8.9–78 fps. Fastest: NVIDIA A100 (server) 833.3 fps (class C), Jetson Thor T5000 382 fps (class D), Jetson AGX Orin 64GB 342 fps (class D).
Change hardware, precision, runtime or add a comparison set on the live engine.
Expected range per precision
Min / median / max of the matching rows, all classes. When a precision has no row, the estimator's benchmark hierarchy (B → C → D) supplies a fallback with its tier chain reported.
| Precision | Metric | N | Min | Median | Max | Best class | Measured |
|---|---|---|---|---|---|---|---|
| INT8 | fps | 2 | 8.9 | 43.4 | 78 | C | 1 |
Rows
| Precision | Runtime | Input | Power mode | Software | Metric | Class | Date | Source | Derivation | Dupes |
|---|---|---|---|---|---|---|---|---|---|---|
| INT8 | hailo sdk (inferred) | 640x640 b8 | default | — | 78 fps | C | 2024-Q1 | Hailo · github:hailo-ai/hailo_model_zoo/HAILO8_object_detection · verified 2026-09-07 | 1 | |
| INT8 | hailo sdk | 640x640 b1 | default | sdk v1 | 8.875 fps | D | 2026 | EdgeAIStack · edgeaistack:benchmark-corpus · YOLOv8 Hailo Research · verified 2026-09-07 | YOLOv8 Hailo Research | 1 |
Cross-hardware comparison
Best row per hardware for the same model and precision, relative to the fastest.
| Hardware | Best value | Class |
|---|---|---|
| NVIDIA A100 (server) (reference) | 833.3 fps | C |
| Jetson Thor T5000 | 382 fps | D |
| Jetson AGX Orin 64GB | 342 fps | D |
| Jetson AGX Orin 32GB | 254 fps | D |
| Jetson Orin Nano Super | 233 fps | D |
| Hailo-10H | 168 fps | D |
| Jetson Orin NX 16GB | 133 fps | D |
| Jetson Orin NX 8GB | 113 fps | D |
| Hailo-8 | 78 fps | C |
| Hailo-8L | 42 fps | C |
| Rockchip RK3588 NPU | 36.5 fps | D |
| Jetson Orin Nano 8GB | 30 fps | D |
| Neousys Nuvo-9531 (Intel 12th-14th Gen Core) | 10 fps | D |
| Raspberry Pi 5 (CPU) (reference) | 5.8 fps | D |
Assumptions
- Rows from the research corpus do not state a runtime; the platform's native runtime (TensorRT / Edge TPU / Hailo SDK / RKNN / CPU) is assumed and flagged runtime_inferred.
- Class C = measured and published by the source named on the row; class D = EdgeAIStack-derived (derivation stated). Rows are inference-only unless scope says end_to_end; publisher measurement conditions (JetPack, batch, warm-up) are in the linked page.
Warnings
- 1 of 2 matching rows are EdgeAIStack-derived (class D); treat them as estimates until a measured row (class C) or your own measurement replaces them.
Method and limitations
Rows are read from the normalised benchmark database (publisher benchmarks and research-derived rows, deduplicated per hardware × model × runtime × precision × input × power mode, best evidence kept). Expected range = min / median / max of the matching rows per precision; when a precision has no row the estimator's benchmark hierarchy (B → C → D) is consulted and its tier chain reported. Comparison = best row per hardware for the same model and precision, relative to the fastest. Nothing is scored; every number carries its class and source line.
Full method, dedupe rule, evidence classes and the confidence rule: Benchmark Explorer methodology.
Measure it yourself.
Nothing measured for your exact config, or want to confirm a class-D row? Run the measurement protocol on your board and submit the result — a human checks it against the protocol before it becomes a class-C row with your attribution.
Links
- Live engine (this exact query): https://edgeaistack.ai/engines/benchmark-explorer/?hw=hailo_8&model=yolov8&variant=s&mv=1.0&dv=2026-09-07
- Methodology: /methodology/benchmark-explorer/
- Measurement protocol: /methodology/measurement-protocol/
- Dataset: /datasets/benchmarks-v2.json
- Hailo-8 overview: /benchmarks/hailo-8/
Change one input and re-run.
The live engine keeps every source line and gives you a fresh permalink and summary.