Phones    |    Mobile SoCs    |    IoT Performance Ranking Desktop GPUs and CPUs
Processor CPU Cores AI Accelerator Year Lib CPU-Q
Score
CPU-F
Score
INT8
NNAPI 1.1
INT8
NNAPI 1.2
INT8
Accuracy
FP16
NNAPI 1.1
FP16
NNAPI 1.2
FP16
Accuracy
INT8
Parallel
FP16
Parallel
LSTM AI Score, K
Qualcomm QCS6052x2.5 GHz Kryo 300/G & 6x1.7 GHz Kryo 300/SDSP (Hex. 685) + GPU (Adreno 615)2018hg109934875846450190.34542684754.67557521232 25.7
i.MX 8M Plus4x2.0 GHz Cortex-A53NPU (Vivante VIP8000)2020nc4881469707295694.417721760.9320509 10.3
Synaptics VideoSmart VS6804x2.3 GHz Cortex-A73NPU (VeriSilicon)2020nc49314976048123094.516223960.92224148 9.6
MediaTek i5004x2.0 GHz Cortex-A73 & 4x2.0 GHz Cortex-A53APU + GPU (Mali-G72 MP3)2019nn87126079064891001207135242.215024027 6.6
Amlogic A311D4x2.2 GHz Cortex-A73 & 2x1.8 GHz Cortex-A55none2018cc802244218928998.84607991001550687 5.8
Amlogic S922X4x1.8 GHz Cortex-A73 & 2x1.9 GHz Cortex-A53none2018cc778232718828798.84417731001548637 5.6
Amlogic S922X4x1.8 GHz Cortex-A73 & 2x1.9 GHz Cortex-A53GPU (Mali-G52 MP4, OpenGL)2018cg779231130728898.8564106485.6163844 5.3
Broadcom BCM27114x1.5 GHz Cortex-A72none2019cc6451946424377973646411003668563 5.1
Rockchip RK35664x1.8 GHz Cortex-A55none2020cc942132759350896.83515941004461423 4.9
JLQ JA3104x1.5 GHz Cortex-A55none2020cc855143652548596.83446241004057422 4.8
Rockchip RK32884x1.8 GHz Cortex-A17none2014cc41710752972621003707231002532494 3.8
Amlogic S905D4x2.0 GHz Cortex-A53none2016cc28510332231901004227651002037398 3.5
Amlogic S905X24x1.8 GHz Cortex-A53none2018cc329125723720398.82263981002430272 3.0
Amlogic S905X34x1.9 GHz Cortex-A55none2019cc342124627222998.82173721002530238 3.0
Amlogic S905X44x2.0 GHz Cortex-A53none2019cc3141155235206972243971002345337 3.0
Hisilicon Hi3798Mv2004x1.6 GHz Cortex-A53none2017cc2218011741501003276291002041316 2.8
Amlogic S905X4x1.5 GHz Cortex-A53none2016cc2117761631321002905091001729340 2.6
Amlogic S905W4x1.5 GHz Cortex-A53none2017cc2187561631441002824991001825326 2.5
Rockchip RK33992x1.8 GHz Cortex-A72 & 4x1.4 GHz Cortex-A55none2016cc4049587312198.8220377100534282 2.5
Amlogic S9054x1.5 GHz Cortex-A53none2015cc2097651331301002415211001729283 2.4
Amlogic S9124x1.5 GHz Cortex-A53 & 4x1.0 GHz Cortex-A53none2016cc27388221418798.81743321001920208 2.3
Rockchip RK33184x1.5 GHz Cortex-A53none2019cc27878018717096.81673061002132226 2.2
Synaptics VideoSmart BG5CT4x1.6 GHz Cortex-A53none2017cc29211362221669711216486.210617 2.1
Allwinner H6164x2.0 GHz Cortex-A53none2020cc225714180151971452621001628225 2.0
Rockchip PX304x1.5 GHz Cortex-A35none2018cc15042511398971061821001423190 1.3
Allwinner T88x1.6 GHz Cortex-A7none2016cc13833299781001402151002027136 1.2
AutoChips AC8227L4x1.1 GHz Cortex-A7none2018cc12232290781001112091001012146 1.1
Allwinner T34x1.2 GHz Cortex-A7none2016cc11931483621001201871001215126 1.1
NXP i.MX 8QuadXPlus4x1.2 GHz Cortex-A35none2017cc1974551446595.3788585.2151435 1.1

xy  -  This device accelerates quantized [x] and floating-point [y] models using the:


n   -   Android NNAPI

g   -   TensorFlow Lite GPU Delegate  (OpenCL / OpenGL based)

h   -   Qualcomm Hexagon NN Direct Delegate

e   -   Samsung Eden Delegate

m  -   MediaTek Neuron Delegate

c   -   TensorFlow Lite / NNAPI default CPU backend


1  -  This SoC might be using unofficial / prototype hardware or drivers

2  -  These are the results of an early prototype. The results of the commercial SoC might be different

Copyright © 2021 by A.I.

ETH Zurich, Switzerland