NVIDIA B300 vs NVIDIA CMP 170HX 10 GB

Comparison of NVIDIA B300 with 144 GB HBM3e and 18,944 cores vs NVIDIA CMP 170HX 10 GB with 10 GB HBM2e and 4,480 cores.

Loading...

Performance Rating

NVIDIA B300 outperforms NVIDIA CMP 170HX 10 GB by 408.34% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA B300

71.9

NVIDIA B300

71.9
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA CMP 170HX 10 GB

14.2

NVIDIA CMP 170HX 10 GB

14.2

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Additional

Memory

Memory Size

🔥 +1,340% 144 ГБ ×2 (288 ГБ)
10 GB

Memory Type

HBM3e HBM2e

Memory Bandwidth

🔥 +163% 4.10 TB/s ×2 (8.2 TB/s)
1.56 TB/s

Memory Bus Width

4,096 бит ×2 (8192 бит) 5,120 бит

ML Performance

FP16 (Half Precision)

1.2318 TFLOPS
🔥 +4,002% 50.53 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +510% 76.99 TFLOPS
12.63 TFLOPS

FP64 (Double Precision)

1.2029 TFLOPS
🔥 +425% 6.317 TFLOPS

CUDA Cores

🔥 +323% 18,944 ×2 (37888)
4,480

RT Cores

No No

Architecture & Compatibility

GPU Architecture

Blackwell Ultra Ampere

SM (Streaming Multiprocessor)

🔥 +111% 148
70

PCIe Version

PCIe 5.0 x16 PCIe 1.0 x4

ML Software Support

CUDA Version

🔥 10.3
8.0

Clocks & Performance

Base Clock

🔥 +46% 1,665
1,140

Boost Clock

🔥 +44% 2,032
1,410

Memory Clock

🔥 +65% 2,000
1,215

Power Consumption

Recommended PSU

1800 W
🔥 -67% 600 W

Power Connector

No 2x 8-pin

TDP/TGP

1400 W
🔥 -82% 250 W

Rendering

Texture Units (TMU)

🔥 +111% 592 ×2 (1184)
280

ROP

No No

L2 Cache

🔥 +400% 50 MB
10 MB

Additional

Slots

🔥 SXM Module
IGP

Release Date

Sept. 11, 2025 Sept. 1, 2021

Display Outputs

No outputs
No outputs

Renting is cheaper than buying