NVIDIA GA10B 28SM vs Qualcomm Adreno X2-85 5.2T Full Specs

Summary

1,792 Shaders 930MHz	1,536 Shaders 1.7GHz
Shared Memory204.8GB/sLPDDR5	Shared Memory152.4GB/sLPDDR5X
Peak·FP32·FP16 3.33 TFLOPS	Peak·FP32·FP16 5.22 TFLOPS
Form Factor iGPU	Form Factor iGPU
TDP Shared	TDP 15W

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP16-16 FP16-16 Tensor Sparse Tensor FP16-32 FP16-32 Tensor Sparse FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

GA10B 28SM213.3 TOPSINT4 Tensor Sparse

x20.42

Adreno X2-85 5.2T10.44 TFLOPSFP16

x1

Show Top Score

Theoretical Performance

Clock Speed930MHz UC·Boost·OC	Clock Speed1700MHz UC·Boost·OC
Peak OPS 213.3 TOPSINT4 Tensor Sparse	Peak OPS 10.44 TFLOPSFP16
Tensor FP16-16 26.66 TFLOPS FP16-16 Tensor Sparse 53.33 TFLOPS Tensor FP16-32 26.66 TFLOPS FP16-32 Tensor Sparse 53.33 TFLOPS	-
BF16 - Tensor BF16 26.66 TFLOPS BF16 Tensor Sparse 53.33 TFLOPS	BF16 10.44 TFLOPS Tensor BF16 - BF16 Tensor Sparse -
Tensor TF32 13.33 TFLOPS	Tensor TF32 -
FP32 3.33 TFLOPS	FP32 5.22 TFLOPS
-	FP64 1.31 TFLOPS
Tensor INT4 106.7 TOPS	Tensor INT4 -
Tensor INT8 53.33 TOPS	-
Pixel Rate 22.3 GPixel/s	Pixel Rate 81.6 GPixel/s
Texture Rate 13 GTexel/s	Texture Rate 163.2 GTexel/s

Core Config

Shaders 1,792 Shaders	Shaders 1,536 Shaders
TMUs 14 TMUs	TMUs 96 TMUs
ROPs 24 ROPs	ROPs 48 ROPs
Tensor Cores 56 T-Cores	Tensor Cores -
SMs 14 SMs	EUs 12 EUs

Clocks

Boost Clock 930MHz

Boost Clock 1.7GHz

Cache

L2 Cache 4.1MB shared	L2 Cache 2MB shared
L3 Cache -	L3 Cache 21MB shared

Memory

Shared MemoryLPDDR5	Shared MemoryLPDDR5X
Memory Bus 256-bit	Memory Bus 128-bit
Memory Speed 6.4GT/s	Memory Speed 9.5GT/s
Memory Bandwidth 204.8GB/s	Memory Bandwidth 152.4GB/s
ECC No	ECC No

Power & Thermals

TDP Shared

TDP 15W

Video Output

Multi-Monitor -

Multi-Monitor 3

Encoder

Encoder Model NVENC 7

Encoder Model Hexagon AV1

Decoder

Decoder Model NVDEC 5

Decoder Model Hexagon AV1

Card

Form Factor iGPU

Form Factor iGPU

Chip

Manufacturer NVIDIA	Manufacturer Qualcomm
Chip Designer NVIDIA	Chip Designer Qualcomm
Architecture Ampere	Architecture Adreno X
Family Jetson	Family Adreno X
Branding	Branding
Codename Hercules	Codename -
Market Segment Embedded	Market Segment Laptop
Release Date Oct 1, 2022	Release Date Sep 24, 2025

Fabrication

Foundry Samsung	Foundry TSMC
Fabrication Node 8N	Fabrication Node N3E
Die Size -	Die Size 288mm²
Transistor Count 17 Billion	Transistor Count -

Images

No images available

No images available

Used In

CPUs AGX Orin 32GB

CPUs Snapdragon X2 Elite (X2E-80-100)

Competitors

NVIDIAGA10B 28SM vs GPU QualcommAdreno 840 NVIDIAGA10B 28SM vs GPU Apple10-Core M5 NVIDIAGA10B 28SM vs GPU Apple10-Core M5-FL Show 11 more

QualcommAdreno X2-85 5.2T vs GPU QualcommAdreno X2-90 7.6T QualcommAdreno X2-85 5.2T vs GPU QualcommAdreno X2-90 7T QualcommAdreno X2-85 5.2T vs GPU IntelArc B370 2400MHz Show 12 more