AMD Radeon AI PRO R9700 vs NVIDIA T40 Full Specs

Summary

4,096 Shaders 2.92GHz	4,608 Shaders 1.56GHz
32GB GDDR6640GB/s	24GB GDDR6624GB/s
Peak·FP32·FP16 47.84 TFLOPS	Peak·FP32·FP16 14.38 TFLOPS
Form Factor PCIe Card	Form Factor PCIe Card
TDP 300W	TDP 150W
Power Connectors 1x 12-Pin - -	Power Connectors - 1x 8-Pin 1x 6-Pin

Highlights

Peak OPS

Peak OPS FP32 FP16 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 Tensor INT4 Memory Bandwidth TDP

Radeon AI PRO R97001.53 POPSINT4 Tensor Sparse

x3.33

T40460.1 TOPSINT4 Tensor

x1

Show Top Score

Theoretical Performance

Clock Speed2920MHz UC·Base·Boost·OC	Clock Speed1560MHz UC·Base·Boost·OC
Peak OPS 1.53 POPSINT4 Tensor Sparse	Peak OPS 460.1 TOPSINT4 Tensor
Tensor FP8-16 191.4 TFLOPS FP8-16 Tensor Sparse 382.7 TFLOPS Tensor FP8-32 191.4 TFLOPS FP8-32 Tensor Sparse 382.7 TFLOPS	-
Tensor FP16-16 95.68 TFLOPS FP16-16 Tensor Sparse 191.4 TFLOPS Tensor FP16-32 95.68 TFLOPS FP16-32 Tensor Sparse 191.4 TFLOPS	Tensor FP16-16 115 TFLOPS FP16-16 Tensor Sparse - Tensor FP16-32 57.51 TFLOPS FP16-32 Tensor Sparse -
BF16 95.68 TFLOPS Tensor BF16 95.68 TFLOPS BF16 Tensor Sparse 191.4 TFLOPS	BF16 - Tensor BF16 57.51 TFLOPS BF16 Tensor Sparse -
Tensor TF32 -	Tensor TF32 57.51 TFLOPS
FP32 47.84 TFLOPS	FP32 14.38 TFLOPS
FP64 1.5 TFLOPS	FP64 449.3 GFLOPS
Tensor INT4 765.5 TOPS	Tensor INT4 460.1 TOPS
Tensor INT8 382.7 TOPS	Tensor INT8 230 TOPS
Ray -	Ray 43.37 TOPS
Pixel Rate 373.8 GPixel/s	Pixel Rate 149.8 GPixel/s
Texture Rate 747.5 GTexel/s	Texture Rate 449.3 GTexel/s

Core Config

Shaders 4,096 Shaders	Shaders 4,608 Shaders
TMUs 256 TMUs	TMUs 288 TMUs
ROPs 128 ROPs	ROPs 96 ROPs
Tensor Cores 128 T-Cores	Tensor Cores 576 T-Cores
RT Cores 64 RT-Cores	RT Cores 72 RT-Cores
CUs 64 CUs	SMs 72 SMs

Clocks

Base Clock 2.4GHz	Base Clock 1.3GHz
Boost Clock 2.92GHz	Boost Clock 1.56GHz

Cache

L2 Cache 8.2MB shared	L2 Cache 6.1MB shared
L3 Cache 64MB shared	L3 Cache -
L3 Bandwidth 2.25TB/s	L3 Bandwidth -

Memory

32GB GDDR6	24GB GDDR6
Memory Bus 256-bit	Memory Bus 384-bit
Memory Speed 20GT/s	Memory Speed 13GT/s
Memory Bandwidth 640GB/s	Memory Bandwidth 624GB/s
ECC No	ECC No

Power & Thermals

TDP 300W

TDP 150W

Video Output

Multi-Monitor 3	Multi-Monitor -
HDCP HDCP 2.3	HDCP -

Ports

3x DisplayPort 2.1	-
1x HDMI 2.1	-

Encoder

Encoder Model VCN 4.0

Encoder Model NVENC 7

Decoder

Decoder Model VCN 4.0

Decoder Model NVDEC 4

Card

Form Factor PCIe Card	Form Factor PCIe Card
PCIe 2-Slots	PCIe 1-Slots
Height 111 mm (4.37") Width 241 mm (9.49") Depth 40 mm (1.57")	Height 111 mm (4.37") Width 267 mm (10.51") Depth 20 mm (0.79")
Cooling Blower 1x Fan	Cooling Passive -
Power Connectors 1x 12-Pin	Power Connectors 1x 8-Pin, 1x 6-Pin

Chip

Manufacturer AMD	Manufacturer NVIDIA
Chip Designer AMD	Chip Designer NVIDIA
Architecture RDNA 4	Architecture Turing
Family Radeon AI PRO	Family Server
Branding	Branding
Codename -	Codename NV162
Chip Variant Navi 48 XL	Chip Variant TU102-300A-K1-A1
Market Segment Workstation	Market Segment Server
Release Date May 20, 2025	Release Date Sep 13, 2018

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node N4P	Fabrication Node 12FFN
Die Size 357mm²	Die Size 754mm²
Transistor Count 53.9 Billion	Transistor Count 18.6 Billion
Transistor Density 151.2 MTr/mm²	Transistor Density 24.67 MTr/mm²

Images

No images available

No images available

Competitors

AMDRadeon AI PRO R9700 vs GPU NVIDIARTX PRO 4500 Blackwell AMDRadeon AI PRO R9700 vs GPU AMDRadeon RX 9070 XT AMDRadeon AI PRO R9700 vs GPU NVIDIARTX 5000 Ada Show 12 more

NVIDIAT40 vs GPU AMDRadeon PRO W7700 NVIDIAT40 vs GPU AMDRadeon RX 9070 XT NVIDIAT40 vs GPU AMDRadeon RX 7800 XT Show 12 more