The 160 W stored here is total board power; NVIDIA lists total graphics power separately at 125 W, the difference being the VirtualLink USB-C connector. Unlike every other Quadro RTX card this datasheet has no ECC row and no NVLink row, so both are recorded as absent rather than assumed. INT8 and INT4 are refused: the Turing whitepaper's TU106 table covers only the GeForce RTX 2070 and NVIDIA never published integer tensor figures for this SKU.

NVIDIA Quadro RTX 4000 8GB

Turing PCIe 2018 12nm Spec confidence: Official
FP32
7.1
TFLOPS
VRAM
8 GB
GDDR6
Bandwidth
416 GB/s
memory
TDP
160 W
44.4 TFLOPS/kW

Overview

The NVIDIA Quadro RTX 4000 is the entry Turing workstation card and the only single-slot member of the family, with 2,304 CUDA cores, 288 second-generation Tensor cores and 36 RT cores, and 8 GB of GDDR6 on a 256-bit bus at 416 GB/s. It delivers 7.1 TFLOPS of FP32 and 57.0 TFLOPS of FP16 tensor throughput inside 160 W. Its appeal was always density and price rather than raw performance: a single slot at 160 W meant it went into workstations and render nodes that could not take a dual-slot board. Eight gigabytes is limiting for AI work today, and Turing predates BF16, TF32, FP8 and structured sparsity, so the 57.0 TFLOPS figure is a dense number with no sparse counterpart. It is included here because it still appears in second-hand and rental supply and needed somewhere accurate to land.

Performance

Peak theoretical throughput by precision type

PrecisionPeak
FP64
64-bit floating point
0.2TFLOPS
FP32
32-bit floating point
7.1TFLOPS
TF32
No verified data available
BF16
No verified data available
FP16
16-bit floating point
57TFLOPS
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.

The Quadro RTX 4000 8GB in the GPU landscape

Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper

06001,2001,8002,4003,0000 W250 W500 W750 W1000 W1250 W1500 WInstinct MI355XQuadro RTX 4000 8GB

Higher and further left is better: more half-precision throughput for less power.

Specifications

Architecture

Turing

Form Factor

PCIe

Launch Year

2018

Process Node

12nm

Memory

8 GB GDDR6

Bandwidth

416 GB/s

TDP

160 W

Max Power

160 W

Transistors

10.8 billion

CUDA Cores

2,304

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 2,304
Tensor Cores 288 (Turing)
Streaming Multiprocessors 36
Chip Design
Transistors 10.8 billion
Die Size 445 mm²
Process Node 12nm
Memory
VRAM 8 GB
Memory Type GDDR6
Bandwidth 416 GB/s
Interface Width 256-bit
Interconnect & I/O
PCIe 3.0 x16
Cache
L2 Cache 4 MB
Power & Thermal
TDP 160 W
Max Board Power 160 W
Cooling Active
Enterprise Features
ECC Memory No
Sparsity No
Compute APIs CUDA, DirectCompute, OpenCL
Physical & Media
Card Length 241.3 mm
Width 1-slot
General
Form Factor PCIe
Architecture Turing
Process Node 12nm
Launch Year 2018

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
2019-04-01

Data Quality

Spec confidence
Official
Clock basis
Boost
Core precisions with figures
3 of 9
Normalization
All values in TFLOPS

Rent this GPU

$0.07 / GPU-hr

Lowest of 1 live listing across 1 provider

See all rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

Similar GPUs

Frequently Asked Questions

How many TFLOPS does the NVIDIA Quadro RTX 4000 have?

The NVIDIA Quadro RTX 4000 delivers 7.1 TFLOPS FP32 and 57 TFLOPS FP16 at peak. Flopper does not currently have a verified FP8 throughput figure for it.

What is the power consumption of the NVIDIA Quadro RTX 4000?

The NVIDIA Quadro RTX 4000 has a TDP (Thermal Design Power) rating of 160 watts.

How much memory does the NVIDIA Quadro RTX 4000 have?

The NVIDIA Quadro RTX 4000 is equipped with 8 GB of VRAM with 416 GB/s of memory bandwidth.

What architecture is the NVIDIA Quadro RTX 4000 based on?

The NVIDIA Quadro RTX 4000 is based on the Turing architecture, launched in 2018.

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