NVIDIA B300 vs Samsung RTX 4070 Max-Q

Comparison of NVIDIA B300 with 144 GB HBM3e and 18,944 cores vs Samsung RTX 4070 Max-Q with 8 GB GDDR6 and 4,608 cores.

Loading...

Performance Rating

NVIDIA B300 outperforms Samsung RTX 4070 Max-Q by 1,058.29% 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

Samsung RTX 4070 Max-Q

6.2

Samsung RTX 4070 Max-Q

6.2

Contents:

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

Memory

Memory Size

🔥 +1,700% 144 ГБ ×2 (288 ГБ)
8 GB

Memory Type

HBM3e GDDR6

Memory Bandwidth

🔥 4.10 TB/s ×2 (8.2 TB/s)
256.0 GB/s

Memory Bus Width

4,096 бит ×2 (8192 бит) 128 бит

ML Performance

FP16 (Half Precision)

1.2318 TFLOPS
🔥 +821% 11.34 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +579% 76.99 TFLOPS
11.34 TFLOPS

FP64 (Double Precision)

🔥 +579% 1.2029 TFLOPS
0.1771 TFLOPS

CUDA Cores

🔥 +311% 18,944 ×2 (37888)
4,608

RT Cores

No
🔥 36

Architecture & Compatibility

GPU Architecture

Blackwell Ultra Ada Lovelace

SM (Streaming Multiprocessor)

🔥 +311% 148
36

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x8

ML Software Support

CUDA Version

🔥 10.3
8.9

Clocks & Performance

Base Clock

🔥 +127% 1,665
735

Boost Clock

🔥 +65% 2,032
1,230

Memory Clock

2,000 2,000

Power Consumption

Recommended PSU

1800 W No

Power Connector

No None

TDP/TGP

1400 W
🔥 -98% 35 W

Rendering

Texture Units (TMU)

🔥 +311% 592 ×2 (1184)
144

ROP

No
🔥 36

L2 Cache

🔥 +56% 50 MB
32 MB

Additional

Slots

🔥 SXM Module
IGP

Release Date

Sept. 11, 2025 Jan. 3, 2023

Display Outputs

No outputs
Portable Device Dependent

Renting is cheaper than buying