NVIDIA B200 vs NVIDIA GB200 Full Specs

Summary

18,944 Shaders 1.98GHz	37,888 Shaders 2.11GHz
192GB HBM3e7.7TB/s	384GB HBM3e16TB/s
Peak·FP32·FP16 75.02 TFLOPS	Peak·FP32·FP16 160 TFLOPS
Form Factor SXM6	Form Factor Superchip
TDP 1000W	TDP 2700W

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 Tensor FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Memory Bandwidth TDP

B20018 PFLOPSFP4 Tensor Sparse

x1

GB20040 PFLOPSFP4 Tensor Sparse (2x 20 PFLOPS)

x2.22

Show Top Score

Theoretical Performance

Clock Speed1980MHz UC·Boost·OC	Clock Speed2112MHz UC·Boost·OC
Peak OPS 18 PFLOPSFP4 Tensor Sparse	Peak OPS 40 PFLOPSFP4 Tensor Sparse
Tensor FP4 9 PFLOPS FP4 Tensor Sparse 18 PFLOPS	Tensor FP4 20 PFLOPS FP4 Tensor Sparse 40 PFLOPS
Tensor FP8-16 4.5 PFLOPS FP8-16 Tensor Sparse 9 PFLOPS Tensor FP8-32 4.5 PFLOPS FP8-32 Tensor Sparse 9 PFLOPS	Tensor FP8-16 10 PFLOPS FP8-16 Tensor Sparse 20 PFLOPS Tensor FP8-32 10 PFLOPS FP8-32 Tensor Sparse 20 PFLOPS
Tensor FP16-16 2.25 PFLOPS FP16-16 Tensor Sparse 4.5 PFLOPS Tensor FP16-32 2.25 PFLOPS FP16-32 Tensor Sparse 4.5 PFLOPS	Tensor FP16-16 5 PFLOPS FP16-16 Tensor Sparse 10 PFLOPS Tensor FP16-32 5 PFLOPS FP16-32 Tensor Sparse 10 PFLOPS
BF16 150 TFLOPS Tensor BF16 2.25 PFLOPS BF16 Tensor Sparse 4.5 PFLOPS	BF16 320.1 TFLOPS Tensor BF16 5 PFLOPS BF16 Tensor Sparse 10 PFLOPS
Tensor TF32 1.13 PFLOPS	Tensor TF32 2.5 PFLOPS
FP32 75.02 TFLOPS	FP32 160 TFLOPS
FP64 37.51 TFLOPS Tensor FP64 35.16 TFLOPS	FP64 80.02 TFLOPS Tensor FP64 78.13 TFLOPS
Tensor INT8 4.5 POPS	Tensor INT8 10 POPS
Pixel Rate 63.4 GPixel/s	Pixel Rate 67.6 GPixel/s
Texture Rate 1.17 TTexel/s	Texture Rate 1.25 TTexel/s

Core Config

Shaders 18,944 Shaders	Shaders 37,888 Shaders
TMUs 592 TMUs	TMUs 1,184 TMUs
ROPs 32 ROPs	ROPs 64 ROPs
Tensor Cores 592 T-Cores	Tensor Cores 1,184 T-Cores
SMs 148 SMs	SMs 296 SMs

Clocks

Boost Clock 1.98GHz	Boost Clock 2.11GHz
Tensor Clock 1.86GHz	Tensor Clock 2.06GHz

Cache

L2 Cache 65.5MB shared

L2 Cache 65.5MB shared

Memory

192GB HBM3e	384GB HBM3e
Memory Bus 8192-bit	Memory Bus 8192-bit
Memory Speed 7.5GT/s	Memory Speed 7.8GT/s
Memory Bandwidth 7.7TB/s	Memory Bandwidth 16TB/s

Power & Thermals

TDP 1000W

TDP 2700W

Decoder

Decoder Model 7x NVDEC 6

Decoder Model 7x NVDEC 6

Card

Form Factor SXM6	Form Factor Superchip
Cooling Passive	Cooling Passive

Chip

Manufacturer NVIDIA	Manufacturer NVIDIA
Chip Designer NVIDIA	Chip Designer NVIDIA
Architecture Blackwell	Architecture Blackwell
Family Server	Family Server
Branding	Branding
Codename NV190	Codename NV190
Chip Variant Umbriel	Chip Variant Umbriel
Market Segment Server	Market Segment Server
Release Date Mar 18, 2024	Release Date Mar 18, 2024

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node 4NP	Fabrication Node 4NP
Die Size 1620mm²	Die Size 1620mm²
Transistor Count 208 Billion	Transistor Count 208 Billion
Transistor Density 128.4 MTr/mm²	Transistor Density 128.4 MTr/mm²

Images

No images available

No images available

Competitors

NVIDIAB200 vs GPU NVIDIAB100 NVIDIAB200 vs GPU AMDInstinct MI350X NVIDIAB200 vs GPU NVIDIAL40 Show 12 more

NVIDIAGB200 vs GPU NVIDIAB100 NVIDIAGB200 vs GPU NVIDIAL40 NVIDIAGB200 vs GPU NVIDIAL40S Show 12 more