Axelera Metis M.2 card: edge AI benchmarks
- Independent test Mario Bergeron
Key results
How to read power: results are compared only within the same model. By efficiency ranks only results whose source publishes a power figure; By speed ranks every result. Measured = measured for this workload. Vendor-measured (typical/avg) = the vendor’s own measurement at typical or average load, not an independent test and not a maximum, so fps/W looks better than at max TDP. Chip, Module or Board TDP = the vendor’s published maximum, not measured. Typical power, not max TDP = only a typical figure is published, so fps/W looks better than at full load. Scope says what the watts cover: Accelerator only (host not included), Module, or Whole system. Max of range = top of a published range, so fps/W is a floor (≥). “— not published” = the source gives no power; we never estimate one.
| Workload | Model · precision | Result | Power | Perf/W | Rank on this model | Grade · date | Source |
|---|---|---|---|---|---|---|---|
| image classification | ResNet-50 v1.5INT8 (vendor quantized) · batch unknown | 2050 FPS | 7.60 WModule Measured | 270● Measured · Module | not in the default fps/W ranking #7 of 50 by speed | C 2026-05-07 (published) | Mario Bergeron, Edge AI Power BenchmarkingMario Bergeron, Edge AI Power Benchmarking (edge-ai-power-p04-axelera-metis), 2026-05-07 |
Ranks compare this board only with boards running the same model. Up to three models per workload; open a rank to see the full same-model table.
Show data — per-workload tables
object detection
No published object detection results for Axelera Metis M.2 card 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 Axelera Metis M.2 card 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 Axelera Metis M.2 card yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
image classification
| Like-for-like group | Best result | Model · precision · stack | Grade | Power | Source · date | Rank in group |
|---|---|---|---|---|---|---|
| mid-size CNN · b=unknown | 2050 FPS | ResNet-50 v1.5 INT8 (vendor quantized) · vendor SDK (see source) | C | 7.60 W ● Measured, Module | Mario Bergeron, Edge AI Power Benchmarking 2026-05-07 (published) | 19 of 47 |
semantic segmentation
No published semantic segmentation results for Axelera Metis M.2 card 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 Axelera Metis M.2 card 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 Axelera Metis M.2 card yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.
Grades: A audited MLPerf · B third-party · C community · D vendor. How we rank · JSON