visual wake words benchmarks on edge AI hardware
MLPerf Tiny, microcontroller class. Throughput in inferences per second; measured energy per inference is shown on each platform page. Not ranked against edge GPUs or accelerators.
35 published rows. 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 (mobile-class CNN, batch 1): Next Generation Snapdragon Mobile Platform MTP at 8753 inferences/s A 2025-09-17 (published). 1 groups, 17 platforms.
Show data — 1 like-for-like tables
mobile-class CNN · batch 1 (17 platforms)
✓ Like-for-like: model-size class mobile-class CNN, batch 1. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Next Generation Snapdragon Mobile Platform MTP | 8753 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) PTQ · Qualcomm AI Stack · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| Snapdragon 8 Gen 3 Mobile HDK (Snapdragon 8 Gen 3 Mobile Platform) | 5211 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) PTQ · Qualcomm AI Stack · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| K1_BOARD_V1.0_HIGH_FREQ (K1 Octa-Core RISC-V) | 458 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform symmetric quantization · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NDP9120 (N/A) | 244 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform scaling from pre-trained weights · N/A · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| K1_BOARD_V1.0 (K1 Octa-Core RISC-V) | 225 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform symmetric quantization · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| EK-RA8D1 (R7FA8D1AHECBD) | 74.67 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-H7A3ZI-Q (STM32H7A3ZIT6Q) | 42.01 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v9.0.0 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-U3C5ZIT6-Q | 26.90 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · STEDGE-AI Internal version · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NUCLEO-H7A3ZI (STM32H7A3ZIT6Q) | 22.85 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| DISCO-F746NG (STM32F746NG) | 17.47 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-U575ZI-Q (STM32U575ZIT6Q) | 16.87 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NUCLEO-L4R5ZI (STM32L4R5ZIT6U) | 11.24 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v9.0 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) | 11.18 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NUCLEO-U385RG-Q (STM32U385RGT6Q) | 10.13 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| LC236R5E176_BOARD_V2.0 (Octopus B1) | 8.31 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform symmetric quantization · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| CY8CPROTO-062-4343W (PSoC 62 MCU) | 6.46 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-G0B1RE (STM32G0B1RE) | 0.707 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
Fastest in the best-covered like-for-like group (mobile-class CNN, batch 1): Next Generation Snapdragon Mobile Platform MTP at 8753 inferences/s A 2025-09-17 (published). 1 groups, 17 platforms.
Show data — 1 like-for-like tables
mobile-class CNN · batch 1 (17 platforms)
✓ Like-for-like: model-size class mobile-class CNN, batch 1. Model, precision and stack may still differ.
| Platform | Result | Model · precision · stack | Grade | Power | Source |
|---|---|---|---|---|---|
| Next Generation Snapdragon Mobile Platform MTP | 8753 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) PTQ · Qualcomm AI Stack · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| Snapdragon 8 Gen 3 Mobile HDK (Snapdragon 8 Gen 3 Mobile Platform) | 5211 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) PTQ · Qualcomm AI Stack · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| K1_BOARD_V1.0_HIGH_FREQ (K1 Octa-Core RISC-V) | 458 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform symmetric quantization · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NDP9120 (N/A) | 244 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform scaling from pre-trained weights · N/A · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| K1_BOARD_V1.0 (K1 Octa-Core RISC-V) | 225 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform symmetric quantization · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| EK-RA8D1 (R7FA8D1AHECBD) | 74.67 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-H7A3ZI-Q (STM32H7A3ZIT6Q) | 42.01 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v9.0.0 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-U3C5ZIT6-Q | 26.90 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · STEDGE-AI Internal version · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NUCLEO-H7A3ZI (STM32H7A3ZIT6Q) | 22.85 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| DISCO-F746NG (STM32F746NG) | 17.47 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-U575ZI-Q (STM32U575ZIT6Q) | 16.87 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NUCLEO-L4R5ZI (STM32L4R5ZIT6U) | 11.24 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v9.0 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) | 11.18 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| NUCLEO-U385RG-Q (STM32U385RGT6Q) | 10.13 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) |
| LC236R5E176_BOARD_V2.0 (Octopus B1) | 8.31 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) Uniform symmetric quantization · · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| CY8CPROTO-062-4343W (PSoC 62 MCU) | 6.46 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
| NUCLEO-G0B1RE (STM32G0B1RE) | 0.707 inferences/s throughput · b=1 | MobileNetV1 0.25 (MLPerf Tiny VWW) not stated · Bosch Hardware-Aware Lowering Engine (HALE) version 1.0.175 · mobile-class CNN | A | unknown power n/a no power published | MLPerf Tiny v1.2 (closed) Older data (2024) |
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.