AMD Viola vs NVIDIA GB10 Full Specs

Summary

3,840 Shaders 2.23GHz	6,144 Shaders 2.52GHz
Shared Memory	Shared Memory273.1GB/sLPDDR5X
Peak·FP32·FP16 34.3 TFLOPS	Peak·FP32·FP16 31.03 TFLOPS
Form Factor iGPU	Form Factor iGPU
TDP 180W	TDP Shared

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP4 FP4 Tensor Sparse Tensor FP8-16 FP8-16 Tensor Sparse Tensor FP8-32 FP8-32 Tensor Sparse 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 Memory Bandwidth TDP

Viola68.6 TFLOPSFP16

x1

GB10992.9 TFLOPSFP4 Tensor Sparse

x14.47

Show Top Score

Theoretical Performance

Clock Speed2233MHz UC·Boost·OC	Clock Speed2525MHz UC·Base·Boost·OC
Peak OPS 68.6 TFLOPSFP16	Peak OPS 992.9 TFLOPSFP4 Tensor Sparse
-	Tensor FP4 496.4 TFLOPS FP4 Tensor Sparse 992.9 TFLOPS
-	Tensor FP8-16 248.2 TFLOPS FP8-16 Tensor Sparse 496.4 TFLOPS Tensor FP8-32 124.1 TFLOPS FP8-32 Tensor Sparse 248.2 TFLOPS
-	Tensor FP16-16 124.1 TFLOPS FP16-16 Tensor Sparse 248.2 TFLOPS Tensor FP16-32 62.05 TFLOPS FP16-32 Tensor Sparse 124.1 TFLOPS
BF16 68.6 TFLOPS Tensor BF16 - BF16 Tensor Sparse -	BF16 62.05 TFLOPS Tensor BF16 62.05 TFLOPS BF16 Tensor Sparse 124.1 TFLOPS
Tensor TF32 -	Tensor TF32 31.03 TFLOPS
FP32 34.3 TFLOPS	FP32 31.03 TFLOPS
FP64 1.07 TFLOPS	FP64 15.51 TFLOPS
-	Tensor INT8 248.2 TOPS
Ray -	Ray 94.1 TOPS
Pixel Rate 214.4 GPixel/s	Pixel Rate 80.8 GPixel/s
Texture Rate 535.9 GTexel/s	Texture Rate 484.8 GTexel/s

Core Config

Shaders 3,840 Shaders	Shaders 6,144 Shaders
TMUs 240 TMUs	TMUs 192 TMUs
ROPs 96 ROPs	ROPs 32 ROPs
Tensor Cores -	Tensor Cores 192 T-Cores
RT Cores 60 RT-Cores	RT Cores 48 RT-Cores
CUs 60 CUs	SMs 48 SMs

Clocks

Base Clock -	Base Clock 2.42GHz
Boost Clock 2.23GHz	Boost Clock 2.52GHz

Cache

L2 Cache 4.1MB shared

L2 Cache 24.6MB shared

Memory

Shared Memory	Shared MemoryLPDDR5X
Memory Bus 256-bit	Memory Bus 256-bit
Memory Speed 18GT/s	Memory Speed 8.5GT/s
Memory Bandwidth 576GB/s	Memory Bandwidth 273.1GB/s

Power & Thermals

TDP 180W

TDP Shared

Video Output

Multi-Monitor 1	Multi-Monitor 4
HDCP -	HDCP HDCP 2.3

Encoder

-	Encoder Model NVENC 9

Decoder

-	Decoder Model NVDEC 6

Card

Form Factor iGPU

Form Factor iGPU

Chip

Manufacturer AMD	Manufacturer NVIDIA
Chip Designer AMD	Chip Designer NVIDIA
Architecture RDNA 4	Architecture Blackwell
Family Console	Family -
Branding	Branding
Codename -	Codename Thor
Market Segment Console	Market Segment Embedded
Release Date Sep 10, 2024	Release Date Jan 6, 2025

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node N6	Fabrication Node N3E
Die Size 260mm²	Die Size -
Transistor Count 10.6 Billion	Transistor Count -
Transistor Density 41 MTr/mm²	Transistor Density -

Images

No images available

No images available

Used In

CPUs Viola	CPUs Grace Blackwell GB10
Desktops -	Desktops DGX Spark(Late 2025)

Competitors

AMDViola vs GPU Apple80-Core M3 Ultra AMDViola vs GPU Apple32-Core M4 Max AMDViola vs GPU NVIDIATegra Switch 2 Show 4 more

NVIDIAGB10 vs GPU Apple60-Core M3 Ultra NVIDIAGB10 vs GPU Apple80-Core M3 Ultra NVIDIAGB10 vs GPU Apple40-Core M4 Max Show 2 more