FP32 6.46 IS DERIVED, NOT PUBLISHED, on the same basis as the GeForce GTX 1070 Ti: 1,920 published CUDA cores times two times the published 1,683 MHz boost clock. NVIDIA printed no TFLOPS figure for this card anywhere, and it never got an announcement of its own -- it was announced inside the GeForce GTX 1080 press release of 2016-05-06, which gives it a price and an availability date and no specifications at all. FP32 IS ALSO THE ONLY PRECISION ROW, for the reason recorded on the GeForce GTX 1080. The archived geforce.com panel labels the two clocks "Graphics Clock 1506" and "Processor Clock 1683" in the site's legacy wording; NVIDIA's later nvidia.com page relabels the identical numbers Base Clock and Boost Clock, which is the mapping used here. REFUSED: FP16, FP64, INT8; l2_cache_mb and ROP count, full-die figures that do not survive harvesting.

NVIDIA GeForce GTX 1070 8GB

Pascal PCIe 2016 16nm Spec confidence: Vendor claimed
FP32
6.5
TFLOPS
Memory
8 GB
GDDR5
Bandwidth
256 GB/s
memory
TDP
150 W
43.1 FP32 dense TFLOPS/kW of TDP
Download datasheet

Overview

The GeForce GTX 1070 was the volume Pascal card of 2016 and it is the cheapest 8 GB NVIDIA GPU that still turns up on GPU marketplaces in numbers. It runs 1,920 CUDA cores at a 1,683 MHz boost for about 6.46 TFLOPS FP32, with 8 GB of GDDR5 at 8 Gbps on a 256-bit bus for 256 GB/s, inside 150 W from a single eight-pin connector. Its appeal for compute was always the frame buffer rather than the arithmetic: 8 GB was generous in 2016 and is still enough for small models, while the throughput is roughly a fifteenth of a modern consumer card once low precision is in play. There are no Tensor Cores, no structured sparsity and no NVIDIA-published half-precision rate anywhere in the Pascal GeForce line, so everything runs in single precision.

Performance

Peak theoretical throughput by precision type

PrecisionPeak
FP64
No verified data available
FP32
32-bit floating point
6.5TFLOPS
TF32
No verified data available
BF16
No verified data available
FP16
No verified data available
FP8
No verified data available
FP6
No verified data available
FP4
No verified data available
INT8
No verified data available

Every figure here is dense. The vendor documents no structured (2:4) sparsity mode for this part, so there is no second number to quote.

Specifications

Architecture

Pascal

Form Factor

PCIe

Launch Year

2016

Process Node

16nm

Memory

8 GB GDDR5

Bandwidth

256 GB/s

TDP

150 W

Max power

150 W

Transistors

7.2 billion

CUDA Cores

1,920

Spec Confidence

Vendor claimed

Full Specifications

Compute Engine
CUDA Cores 1,920
Streaming Multiprocessors 15
Base Clock 1.51 GHz
Boost Clock 1.68 GHz
Chip Design
Transistors 7.2 billion
Die Size 314 mm²
Process Node 16nm
Memory
Memory 8 GB
Memory Type GDDR5
Bandwidth 256 GB/s
Interface Width 256-bit
Memory Clock 8 GT/s
Interconnect & I/O
PCIe 3.0 x16
Power & Thermal
TDP 150 W
Max power 150 W
Power Connector 8-pin
Cooling Active
Enterprise Features
ECC Memory No
Sparsity No
Compute APIs CUDA, DirectX 12, Vulkan, OpenGL
Physical & Media
Card Length 266.7 mm
Width 2-slot
General
Form Factor PCIe
Architecture Pascal
Process Node 16nm
Launch Year 2016

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
No verified data available

Data Quality

Spec confidence
Vendor claimed
Clock basis
Boost
Core precisions with figures
1 of 9
Normalization
All values in TFLOPS

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

Browse GPU Cloud Providers

Similar GPUs

NVIDIA Titan Xp 12GB

PCIe · 2017
FP32: 12 TFLOPS
Compare vs Titan Xp

NVIDIA Tesla P40 24GB

PCIe · 2016
FP32: 12 TFLOPS
Compare vs Tesla P40

NVIDIA P100 SXM2 16GB

SXM · 2016
FP32: 11 TFLOPS
Compare vs P100

NVIDIA P100 PCIe 16GB

PCIe · 2016
FP32: 9.3 TFLOPS
Compare vs P100

Frequently Asked Questions

How many TFLOPS does the NVIDIA GeForce GTX 1070 have?

The NVIDIA GeForce GTX 1070 delivers 6.5 TFLOPS FP32 at peak. Flopper does not currently have verified FP16 and FP8 throughput figures for it.

What is the power consumption of the NVIDIA GeForce GTX 1070?

The NVIDIA GeForce GTX 1070 has a TDP (Thermal Design Power) rating of 150 watts.

How much memory does the NVIDIA GeForce GTX 1070 have?

The NVIDIA GeForce GTX 1070 is equipped with 8 GB of memory with 256 GB/s of memory bandwidth.

What architecture is the NVIDIA GeForce GTX 1070 based on?

The NVIDIA GeForce GTX 1070 is based on the Pascal architecture, launched in 2016.

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