NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P): edge AI benchmarks
Key results
| Workload | Best result | Model · precision | Power · perf/W | Grade · date | Source |
|---|---|---|---|---|---|
| tiny image classification | 6.04 inferences/s 5496 µJ/inference energy (lower is better) batch 1 | ResNet-8 (MLPerf Tiny IC) INT8 · X-CUBE-AI v10.2 | unknown | A 2025-09-17 (published) | MLPerf Tiny v1.3 (closed) |
| visual wake words | 11.18 inferences/s 3030 µJ/inference energy (lower is better) batch 1 | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 | unknown | A 2025-09-17 (published) | MLPerf Tiny v1.3 (closed) |
| anomaly detection | 209 inferences/s 160 µJ/inference energy (lower is better) batch 1 | FC autoencoder (MLPerf Tiny AD) INT8 · X-CUBE-AI v10.2 | unknown | A 2025-09-17 (published) | MLPerf Tiny v1.3 (closed) |
| keyword spotting | 23.03 inferences/s 1431 µJ/inference energy (lower is better) batch 1 | DS-CNN (MLPerf Tiny KWS) INT8 · X-CUBE-AI v10.2 | unknown | A 2025-09-17 (published) | MLPerf Tiny v1.3 (closed) |
Best published result per workload, across model sizes. For like-for-like groups, open the per-workload tables below.
Show data — per-workload tables
object detection
No published object detection results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
instance segmentation
No published instance segmentation results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
pose estimation
No published pose estimation results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
image classification
No published image classification results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
semantic segmentation
No published semantic segmentation results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
speech-to-text
No published speech-to-text results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
LLM generation
No published LLM generation results for NUCLEO-L4R5ZI-P (STM32L4R5ZIT6P) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
tiny image classification
| Like-for-like group | Best result | Model · precision · stack | Grade | Power | Source · date | Rank in group |
|---|---|---|---|---|---|---|
| mid-size CNN · b=1 | 6.04 inferences/s | ResNet-8 (MLPerf Tiny IC) INT8 · X-CUBE-AI v10.2 | A | unknown power n/a | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) | 16 of 19 |
visual wake words
| Like-for-like group | Best result | Model · precision · stack | Grade | Power | Source · date | Rank in group |
|---|---|---|---|---|---|---|
| mobile-class CNN · b=1 | 11.18 inferences/s | MobileNetV1 0.25 (MLPerf Tiny VWW) INT8 · X-CUBE-AI v10.2 | A | unknown power n/a | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) | 13 of 17 |
anomaly detection
| Like-for-like group | Best result | Model · precision · stack | Grade | Power | Source · date | Rank in group |
|---|---|---|---|---|---|---|
| unknown · b=1 | 209 inferences/s | FC autoencoder (MLPerf Tiny AD) INT8 · X-CUBE-AI v10.2 | A | unknown power n/a | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) | 12 of 16 |
keyword spotting
| Like-for-like group | Best result | Model · precision · stack | Grade | Power | Source · date | Rank in group |
|---|---|---|---|---|---|---|
| unknown · b=1 | 23.03 inferences/s | DS-CNN (MLPerf Tiny KWS) INT8 · X-CUBE-AI v10.2 | A | unknown power n/a | MLPerf Tiny v1.3 (closed) 2025-09-17 (published) | 13 of 17 |
How to read these numbers
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.