AMD Instinct MI100 vs AMD Radeon Vega 7 1800MHz Full Specs

Summary

7,680 Shaders 1.5GHz	448 Shaders 1.8GHz
32GB HBM21.23TB/s placeholder	Shared Memory68.3GB/s LPDDR4X·DDR4
Peak·FP32·FP16 23.07 TFLOPS	Peak·FP32·FP16 1.61 TFLOPS
Form Factor PCIe Card	Form Factor iGPU
TDP 300W	TDP Shared
Power Connectors 2x 8-Pin	Power Connectors -

Highlights

Peak OPS

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

Instinct MI100184.6 TFLOPSFP16-16 Tensor

x57.22

Radeon Vega 7 1800MHz3.23 TFLOPSFP16

x1

Show Top Score

Theoretical Performance

Clock Speed1502MHz UC·Base·Boost·OC	Clock Speed1800MHz UC·Base·Boost·OC
Peak OPS 184.6 TFLOPSFP16-16 Tensor	Peak OPS 3.23 TFLOPSFP16
Tensor FP16-16 184.6 TFLOPS Tensor FP16-32 184.6 TFLOPS	-
Tensor BF16 92.28 TFLOPS	-
FP32 23.07 TFLOPS Tensor FP32 46.14 TFLOPS	FP32 1.61 TFLOPS Tensor FP32 -
FP64 11.54 TFLOPS	FP64 100.8 GFLOPS
Tensor INT4 92.28 TOPS	Tensor INT4 -
Tensor INT8 92.28 TOPS	-
Pixel Rate -	Pixel Rate 14.4 GPixel/s
Texture Rate 721 GTexel/s	Texture Rate 50.4 GTexel/s

Core Config

Shaders 7,680 Shaders	Shaders 448 Shaders
TMUs 480 TMUs	TMUs 28 TMUs
ROPs -	ROPs 8 ROPs
Tensor Cores 480 T-Cores	Tensor Cores -
CUs 120 CUs	CUs 7 CUs

Clocks

Base Clock 1GHz	Base Clock 300MHz
Boost Clock 1.5GHz	Boost Clock 1.8GHz

Cache

L1 16KB/CU	L1 16KB/CU
L2 Cache 8MB shared	L2 Cache 1MB shared

Memory

32GB HBM2 placeholder	Shared Memory LPDDR4X·DDR4
Memory Bus 4096-bit	Memory Bus 128-bit
Memory Speed 2.4GT/s	Memory Speed 4.3GT/s
Memory Bandwidth 1.23TB/s	Memory Bandwidth 68.3GB/s
ECC Yes	ECC No

Power & Thermals

TDP 300W

TDP Shared

Video Output

Max Resolution -	Max Resolution 4096x2160
Refresh Rate Calculator -	Refresh Rate Calculator 3840x2160 Resolution4K UHD Refresh Rate 64Hz
Multi-Monitor -	Multi-Monitor 3
Adaptive Sync -	Adaptive Sync FreeSync
DSC Not Supported	DSC Not Supported

API Support

Shader Model -	Shader Model 6.7
-	DirectX DirectX 12 Direct3D 12_1
OpenGL - OpenCL 2.1 Vulkan -	OpenGL 4.6 OpenCL 3 Vulkan 1.2
GFX 9.4	GFX 9

Encoder

Encoder VCN 2.5	Encoder VCN 2.0
Codec AVC (H.264) HEVC (H.265)	Codec AVC (H.264) HEVC (H.265)

Decoder

Decoder VCN 2.5	Decoder VCN 2.0
Codec MPEG-1 MPEG-2 MPEG-4 JPEG VC-1 VP9 AVC (H.264) HEVC (H.265)	Codec MPEG-1 MPEG-2 MPEG-4 JPEG VC-1 VP9 AVC (H.264) HEVC (H.265)

Card

Form Factor PCIe Card	Form Factor iGPU
PCIe PCIe 4.0 x16 2-Slots	PCIe -
Height 111mm (4.37") Width 267mm (10.51") Depth 37mm (1.46")	-
Cooling Passive	Cooling Open-Air
Power Connectors 2x 8-Pin	Power Connectors -
Multi-GPU Multi-GPU Support Supported Multi-GPU Type Infinity Fabric	Multi-GPU -

Chip

Manufacturer AMD	Manufacturer AMD
Chip Designer AMD	Chip Designer AMD
Architecture CDNA 1	Architecture GCN 5
Family Instinct	Family Vega
Branding	Branding
Codename Arcturus	Codename -
Chip Variant Arcturus XL	Chip Variant Vega 3
Market Segment Server	Market Segment Laptop
Release Date Nov 16, 2020	Release Date Jan 12, 2021

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node N7	Fabrication Node N7
Die Size 750mm²	Die Size 180mm²
Transistor Count 25.6B	Transistor Count 10.7B
Transistor Density 34.13 MTr/mm²	Transistor Density 59 MTr/mm²

Images