NVIDIA GeForce RTX 3070 8GB

Ampere PCIe 2020 8nm
FP32
20.3
TFLOPS
VRAM
8 GB
GDDR6
TDP
220 W
92.3 TFLOPS/kW
Bandwidth
448 GB/s
memory
CUDA Cores
5,888
184 tensor cores
View Providers

Performance Metrics

Peak theoretical throughput by precision type

PrecisionDescriptionBitsPeak TFLOPSEfficiency
INT88-bit integer840.60.185 TFLOPS/W
FP1616-bit floating point1620.30.092 TFLOPS/W
BF16Brain Float 161620.30.092 TFLOPS/W
TF32TensorFloat-323220.30.092 TFLOPS/W
FP3232-bit floating point3220.30.092 TFLOPS/W
FP6464-bit floating point640.30.001 TFLOPS/W
FP16 Efficiency
0.092 TFLOPS/W
20.3 TFLOPS / 220W
FP32 Efficiency
0.092 TFLOPS/W
20.3 TFLOPS / 220W

Power Specifications

TDP

220 W

Max Power

253 W

Power Connector

PCIe 8-pin

Cooling

Air

Memory Specifications

Capacity

8 GB

Type

GDDR6

Bandwidth

448 GB/s

Interface

256-bit

Hardware & Design

Form Factor

PCIe

Architecture

Ampere

Process Node

8nm

Launch Year

2020

Variant

8GB

Market Segment

Professional

Transistors

17.4 billion

Die Size

393 mm²

Full Specifications

Compute Engine
CUDA Cores5,888
Tensor Cores184 (3rd Gen)
Streaming Multiprocessors46
Base Clock1.50 GHz
Boost Clock1.73 GHz
Chip Design
Transistors17.4 billion
Die Size393 mm²
Process Node8nm
Memory
VRAM8 GB
Memory TypeGDDR6
Bandwidth448 GB/s
Interface Width256-bit
Power & Thermal
TDP220 W
General
Form FactorPCIe
ArchitectureAmpere
Launch Year2020

Common Use Cases

General Compute AI/ML Workloads Data Processing

The NVIDIA GeForce RTX 3070 is optimized for high-performance computing tasks with Ampere architecture delivering 20.31 TFLOPS of compute power.

Where to Rent

Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.

Browse GPU Cloud Providers

Similar GPUs

NVIDIA RTX A6000 48GB

PCIe · 2020
FP32: 38.7 TFLOPS
Compare vs RTX A6000

NVIDIA A40 48GB

PCIe · 2021
FP32: 37.4 TFLOPS
Compare vs A40

NVIDIA A10G 24GB

PCIe · 2022
FP32: 35.0 TFLOPS
Compare vs A10G

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs