Manli RTX 3070 (M2479+N617-00) vs NVIDIA A100 PCIe 80 GB

Comparison of Manli RTX 3070 (M2479+N617-00) with 8 GB GDDR6 and 5,888 cores vs NVIDIA A100 PCIe 80 GB with 80 GB HBM2e and 6,912 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

Manli RTX 3070 (M2479+N617-00)

10.3

Manli RTX 3070 (M2479+N617-00)

10.3
RX 7900 XTX RX 7900 XTX
Instinct MI300X Instinct MI300X
MI250 MI250

NVIDIA A100 PCIe 80 GB

NVIDIA A100 PCIe 80 GB

Contents:

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

Memory

Memory Size

🔥 8 ГБ
80 GB

Memory Type

GDDR6 HBM2e

Memory Bandwidth

448.0 GB/s 1.94 TB/s

Memory Bus Width

256 бит 5,120 бит

ML Performance

FP16 (Half Precision)

🔥 20.31 TFLOPS
77.97 TFLOPS

BF16 (Brain Float)

No 311.84 TFLOPS

TF32 (TensorFloat)

No 155.92

Compute Power

FP32 (Single Precision)

🔥 +4% 20.31 TFLOPS
19.49 TFLOPS

FP64 (Double Precision)

🔥 0.31739999999999996 TFLOPS
9.746 TFLOPS

CUDA Cores

🔥 5,888
6,912

RT Cores

🔥 46
No

Architecture & Compatibility

GPU Architecture

Ampere Ampere

SM (Streaming Multiprocessor)

🔥 46
108

PCIe Version

PCIe 4.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

🔥 8.6
8.0

Clocks & Performance

Base Clock

🔥 +41% 1,500
1,065

Boost Clock

🔥 +22% 1,725
1,410

Memory Clock

🔥 +16% 1,750
1,512

Power Consumption

TDP/TGP

🔥 -27% 220 W
300 W

Recommended PSU

🔥 -21% 550 W
700 W

Power Connector

1x 12-pin 8-pin EPS

Rendering

Texture Units (TMU)

🔥 184
432

ROP

🔥 46
No

L2 Cache

🔥 4 MB
80 MB

Additional

Slots

Triple-slot Dual-slot

Release Date

Sept. 1, 2020 June 28, 2021

Display Outputs

1x HDMI 2.1
3x DisplayPort 1.4a
No outputs

Renting is cheaper than buying