GPUs

NVIDIA GB10 vs NVIDIA GeForce RTX 4070 Super Full Specs

6,144 Shaders
2.52GHz
7,168 Shaders
2.48GHz
Shared Memory273.1GB/sLPDDR5X
12GB GDDR6X504.2GB/s
··
31.03 TFLOPS
··
35.48 TFLOPS
Form Factor
iGPU
Form Factor
PCIe Card
TDP
Shared
TDP
225W
Power Connectors
-
Power Connectors
1x 16-Pin 12VHPWR

GB10GB10992.9 TFLOPSFP4 Tensor Sparse
x1
GeForce RTX 4070 SuperGeForce RTX 4070 Super1.14 POPSINT4 Tensor Sparse
x1.14

Clock Speed
···
Clock Speed
···
Peak OPS
992.9 TFLOPSFP4 Tensor Sparse
Peak OPS
1.14 POPSINT4 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 FP8-16
283.9 TFLOPS
FP8-16 Tensor Sparse
567.7 TFLOPS
Tensor FP8-32
141.9 TFLOPS
FP8-32 Tensor Sparse
283.9 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
Tensor FP16-16
141.9 TFLOPS
FP16-16 Tensor Sparse
283.9 TFLOPS
Tensor FP16-32
70.96 TFLOPS
FP16-32 Tensor Sparse
141.9 TFLOPS
BF16
62.05 TFLOPS
Tensor BF16
62.05 TFLOPS
BF16 Tensor Sparse
124.1 TFLOPS
BF16
35.48 TFLOPS
Tensor BF16
70.96 TFLOPS
BF16 Tensor Sparse
141.9 TFLOPS
Tensor TF32
31.03 TFLOPS
Tensor TF32
35.48 TFLOPS
FP32
31.03 TFLOPS
FP32
35.48 TFLOPS
FP64
15.51 TFLOPS
FP64
554.4 GFLOPS
Tensor INT4
-
Tensor INT4
567.7 TOPS
Tensor INT8
248.2 TOPS
Tensor INT8
283.9 TOPS
Ray
94.1 TOPS
Ray
82.01 TOPS
Pixel Rate
80.8 GPixel/s
Pixel Rate
198 GPixel/s
Texture Rate
484.8 GTexel/s
Texture Rate
554.4 GTexel/s

Shaders
6,144 Shaders
Shaders
7,168 Shaders
TMUs
192 TMUs
TMUs
224 TMUs
ROPs
32 ROPs
ROPs
80 ROPs
Tensor Cores
192 T-Cores
Tensor Cores
224 T-Cores
RT Cores
48 RT-Cores
RT Cores
56 RT-Cores
SMs
48 SMs
SMs
56 SMs

Base Clock
2.42GHz
Base Clock
1.98GHz
Boost Clock
2.52GHz
Boost Clock
2.48GHz

L2 Cache
24.6MB shared
L2 Cache
36.9MB shared

Shared MemoryLPDDR5X
12GB GDDR6X
Memory Bus
256-bit
Memory Bus
192-bit
Memory Speed
8.5GT/s
Memory Speed
21GT/s
Memory Bandwidth
273.1GB/s
Memory Bandwidth
504.2GB/s
ECC
No
ECC
No

TDP
Shared
TDP
225W

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

-
3x DisplayPort 1.4
-
1x HDMI 2.1

Encoder Model
NVENC 9
Encoder Model
2x NVENC 8

Decoder Model
NVDEC 6
Decoder Model
NVDEC 5

Form Factor
iGPU
Form Factor
PCIe Card
PCIe
-
PCIe
2.1-Slots
-
Height
112 mm (4.41")
Width
267 mm (10.51")
Depth
42 mm (1.65")
Cooling
Open-Air
-
Cooling
Open-Air
2x Fans
Power Connectors
-
Power Connectors
1x 16-Pin 12VHPWR

Manufacturer
Manufacturer
Chip Designer
Chip Designer
Architecture
Architecture
Family
-
Family
Branding
GB10 branding
Branding
GeForce RTX 2022 branding
Codename
Thor
Codename
NV184
Chip Variant
-
Chip Variant
AD104-250
Market Segment
Embedded
Market Segment
Desktop
Release Date
Jan 6, 2025
Release Date
Jan 8, 2024

Foundry
TSMC
Foundry
TSMC
Fabrication Node
N3E
Fabrication Node
4N
Die Size
-
Die Size
295mm²
Transistor Count
-
Transistor Count
35.8 Billion
Transistor Density
-
Transistor Density
121.6 MTr/mm²

No images available
No images available