NVIDIA GB200 Specs

NVIDIA GB200
2x 18944 Shaders
2x 192GB HBM3e
2112MHz
Peak Performance
40000.03 TFLOPS
FP4 Tensor Sparse
FP32
160.04 TFLOPS
FP16
320.08 TFLOPS
Form Factor
Superchip
-
TDP
2700W
-
-
-
-
-
GB6 OpenCL N/A
0%
GB6 Metal N/A
0%
GB6 Vulkan N/A
0%
GB5 OpenCL N/A
0%
GB5 CUDA N/A
0%
GB5 Metal N/A
0%
GB5 Vulkan N/A
0%
OCT 2020.1 N/A
0%
OCT Metal N/A
0%
Manufacturer
NVIDIA
Chip Designer
NVIDIA
Architecture
Blackwell
Family
Server
Codename
NV190
GB100
Variant
Oberon
Market Segment
Server
Release Date
3/18/2024
Foundry
TSMC
-
Fabrication Node
4NP
-
Die Size
4x 810 mm²
-
Transistor Count
4x 104 Billion
-
Transistor Density
128.40M/mm²
-
Form
Superchip
Shading Units
2x 18944 Shaders
-
Texture Mapping Units
2x 592 TMUs
Render Output Units
2x 32 ROPs
Tensor Cores
2x 592 T-Cores
-
-
Streaming Multiprocessors
2x 148 SMs
-
-
-
-
-
-
2062MHz Tensor
-
-
2112MHz
Peak Performance
40000.03 TFLOPS
FP4 Tensor Sparse
FP4
20000.01 TFLOPS Tensor
40000.03 TFLOPS Tensor Sparse
FP8
-
10000.01 TFLOPS Tensor (FP16 Accumulate)
20000.02 TFLOPS Tensor (FP16 Accumulate) Sparse
10000.01 TFLOPS Tensor (FP32 Accumulate)
20000.02 TFLOPS Tensor (FP32 Accumulate) Sparse
FP16
320.08 TFLOPS
5000 TFLOPS Tensor (FP16 Accumulate)
10000.01 TFLOPS Tensor (FP16 Accumulate) Sparse
5000 TFLOPS Tensor (FP32 Accumulate)
10000.01 TFLOPS Tensor (FP32 Accumulate) Sparse
FP32
160.04 TFLOPS
-
-
FP64
80.02 TFLOPS
78.13 TFLOPS Tensor
BF16
320.08 TFLOPS
5000 TFLOPS Tensor
10000.01 TFLOPS Tensor Sparse
TF32
2500 TFLOPS Tensor
5000 TFLOPS Tensor Sparse
-
-
-
INT8
-
10000.01 TOPS Tensor
20000.02 TOPS Tensor Sparse
INT32
160.04 TOPS
-
-
Pixel Fillrate
67.584 GPixel/s
-
-
Texture Fillrate
1250.304 GTexel/s
-
-
L1
64KB/SM Tex
256KB/SM
-
-
L2
64MB Shared
-
-
-
2x 192GB
HBM3e
-
Bus Width
8192Bit
Clock
3907MHz
Transfer Rate
7.8GT/s
Bandwidth
8001.5GB/s
-
-
-
-
-
-
-
-
-
TDP
2700W
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
Max Resolution
Unknown
Max Resolution Refresh Rate
-
Variable Refresh Rate
G-Sync
FreeSync
-
Display Stream Compression (DSC)
Not Supported
Multi Monitor Support
Unknown
-
-
No Encoders
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Model
7x NVDEC 6
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
-
-
-
-
-
-
OpenCL
3.0
-
-
-
-
CUDA
10.0
-
-
PureVideo HD
VP13
VDPAU
Feature Set M
Not a Card
-
-
-
-
-
-
-
-
-
-
-
-
-
PCIe Version
6.0
PCIe Lanes
16
Multi GPU Support
Supported
Type
NVLink
-
-
-
-
-
-
Change Comparison