AMD Versal AI Edge VEK280 (VE2802) vs Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0

FPGA / Versal coverage note: 1,243 rows come from the AMD Vitis AI 3.0/3.5 Model Zoo tables; 733 (59%) are image classification. Detection, segmentation and other workloads are thinner, and there is no LLM or speech data for these boards. All are vendor-published (grade D). FPGA / Versal overview.

AMDeval kit (adaptive SoC)No licensed photo yet

GoogleAI accelerator (USB)No licensed photo yet

Showing the newest results per platform and source. 18 superseded older results are hidden.Showing all results, including 18 superseded by newer results from the same source.

No like-for-like comparison yet: both have published results, but model size or batch differ. Treat the numbers below as indicative only.

	AMD Versal AI Edge VEK280 (VE2802)	Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0
Best semantic segmentation	1564 FPS D Older data (2023)	303 FPS D Older data (2020)
Power	n/a	2.00 W spec
Perf / W	n/a	152 FPS/W spec
Price	n/a	n/a
Workloads with data	4 of 7	1 of 7

Best published semantic segmentation result per platform. n/a = not published; never estimated.

Show data — all workloads, sources and like-for-like groups

object detection

Only AMD Versal AI Edge VEK280 (VE2802) has published object detection results.

Platform	Result	Model · precision · stack	Grade	Power	Source
AMD Versal AI Edge VEK280 (VE2802)	164 FPS E2E throughput, multi thread · b=14	yolov5l_pt INT8 · Vitis AI 3.5 Library (VART), batch 14 DPU · large (YOLO-l)	D	unknown power n/a no power published	AMD UG1354 Vitis AI Library performance tables undated

Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0:

◌

No published object detection results for Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0 or AMD Versal AI Edge VEK280 (VE2802) 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 AMD Versal AI Edge VEK280 (VE2802) or Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0 yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.

pose estimation

Only AMD Versal AI Edge VEK280 (VE2802) has published pose estimation results.

Platform	Result	Model · precision · stack	Grade	Power	Source
AMD Versal AI Edge VEK280 (VE2802)	428 FPS E2E throughput, multi thread · b=unknown	pt_movenet_3.5 INT8 · Vitis AI 3.5 (VART) · mobile-class CNN	D	unknown power n/a no power published	Vitis AI 3.5 Model Zoo performance table Older data (2023)

Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0:

◌

No published pose estimation results for Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0 or AMD Versal AI Edge VEK280 (VE2802) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.

image classification

Only AMD Versal AI Edge VEK280 (VE2802) has published image classification results.

Platform	Result	Model · precision · stack	Grade	Power	Source
AMD Versal AI Edge VEK280 (VE2802)	337 FPS throughput · b=1	resnet50 (TRD example xmodel) INT8 · Vitis AI 3.5 DPUCV2DX8G reference design, test_dpu_runner_mt · mid-size CNN	D	unknown power n/a no power published	Vitis AI 3.5 VEK280 DPUCV2DX8G reference design (ResNet50 example) Older data (2023)

Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0:

◌

No published image classification results for Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0 or AMD Versal AI Edge VEK280 (VE2802) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.

semantic segmentation

⚠ MIXED COMPARISON — not like-for-like. Baseline mid-size CNN at batch 1. Differs: batch unknown vs 1.

Platform	Result	Model · precision · stack	Grade	Power	Source
Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0	303 FPS inference_time · b=1	Unet Mv2 (128x128) INT8 · TensorFlow Lite C++ benchmark (Edge TPU runtime) · mid-size CNN	D	2.00 W spec platform spec envelope (not measured with this result)	Google Coral Edge TPU benchmarks page (Table 1) Older data (2020)
AMD Versal AI Edge VEK280 (VE2802) MIXED: batch unknown vs 1	1564 FPS E2E throughput, multi thread · b=unknown	tf2_2D-UNET_3.5 INT8 · Vitis AI 3.5 (VART) · mid-size CNN	D	unknown power n/a no power published	Vitis AI 3.5 Model Zoo performance table Older data (2023)

No shared like-for-like group exists for this pair.

speech-to-text

◌

No published speech-to-text results for AMD Versal AI Edge VEK280 (VE2802) or Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0 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 AMD Versal AI Edge VEK280 (VE2802) or Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0 yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.

AMD Versal AI Edge VEK280 (VE2802) is faster in the only like-for-like comparison (largest gap: image classification, 16.51×).

	AMD Versal AI Edge VEK280 (VE2802)	Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0
Best image classification	8598 FPS D Older data (2023)	476 FPS D Older data (2019)
Power	n/a	2.00 W spec
Perf / W	n/a	238 FPS/W spec
Price	n/a	n/a
Workloads with data	4 of 7	3 of 7

Best published image classification result per platform (may be from different like-for-like groups — see data). n/a = not published; never estimated.

Show data — all workloads, sources and like-for-like groups

object detection

⚠ MIXED COMPARISON — not like-for-like. Baseline mobile-class CNN at batch 1. Differs: batch unknown vs 1.

Platform	Result	Model · precision · stack	Grade	Power	Source
Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0	154 FPS inference_time · b=1	MobileNet v1 SSD (224x224) INT8 · TensorFlow Lite C++ benchmark (Edge TPU runtime) · mobile-class CNN	D	2.00 W spec platform spec envelope (not measured with this result)	Google Coral Edge TPU benchmarks page (Table 1) Older data (2019)
AMD Versal AI Edge VEK280 (VE2802) MIXED: batch unknown vs 1	1933 FPS E2E throughput, multi thread · b=unknown	tf_ssdmobilenetv1_3.5 INT8 · Vitis AI 3.5 (VART) · mobile-class CNN	D	unknown power n/a no power published	Vitis AI 3.5 Model Zoo performance table Older data (2023)

No shared like-for-like group exists for this pair.

instance segmentation

◌

pose estimation

Only AMD Versal AI Edge VEK280 (VE2802) has published pose estimation results.

Platform	Result	Model · precision · stack	Grade	Power	Source
AMD Versal AI Edge VEK280 (VE2802)	428 FPS E2E throughput, multi thread · b=unknown	pt_movenet_3.5 INT8 · Vitis AI 3.5 (VART) · mobile-class CNN	D	unknown power n/a no power published	Vitis AI 3.5 Model Zoo performance table Older data (2023)

Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0:

◌

image classification

✓ Like-for-like: model-size class mid-size CNN, batch 1. Model, precision and stack may still differ.

Platform	Result	Model · precision · stack	Grade	Power	Source
AMD Versal AI Edge VEK280 (VE2802)	337 FPS throughput · b=1	resnet50 (TRD example xmodel) INT8 · Vitis AI 3.5 DPUCV2DX8G reference design, test_dpu_runner_mt · mid-size CNN	D	unknown power n/a no power published	Vitis AI 3.5 VEK280 DPUCV2DX8G reference design (ResNet50 example) Older data (2023)
Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0	20.41 FPS inference_time · b=1	ResNet-50 V1 (299x299) INT8 · TensorFlow Lite C++ benchmark (Edge TPU runtime) · mid-size CNN	D	2.00 W spec platform spec envelope (not measured with this result)	Google Coral Edge TPU benchmarks page (Table 1) Older data (2019)

semantic segmentation

⚠ MIXED COMPARISON — not like-for-like. Baseline mid-size CNN at batch 1. Differs: batch unknown vs 1.

Platform	Result	Model · precision · stack	Grade	Power	Source
Google Coral USB Accelerator (Edge TPU) on Xeon Gold 6154 host, USB 3.0	303 FPS inference_time · b=1	Unet Mv2 (128x128) INT8 · TensorFlow Lite C++ benchmark (Edge TPU runtime) · mid-size CNN	D	2.00 W spec platform spec envelope (not measured with this result)	Google Coral Edge TPU benchmarks page (Table 1) Older data (2020)
AMD Versal AI Edge VEK280 (VE2802) MIXED: batch unknown vs 1	1564 FPS E2E throughput, multi thread · b=unknown	tf2_2D-UNET_3.5 INT8 · Vitis AI 3.5 (VART) · mid-size CNN	D	unknown power n/a no power published	Vitis AI 3.5 Model Zoo performance table Older data (2023)

No shared like-for-like group exists for this pair.

speech-to-text

◌

LLM generation

◌

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.