NVIDIA publishes two different tensor figures for this card. The datasheet PDF (2895141, SEP23) prints "Tensor Performance 327.6 TFLOPS", while the product page prints 427.6 TFLOPS. 427.6 is the coherent one: it is 16x the card's own 26.7 TFLOPS FP32, which is the ratio NVIDIA's sparse FP8 headline holds on every other Ada part including this card's own SFF sibling (306.8 against 19.2), whereas 327.6 is 12.3x and matches no Ada rate at all. The stored FP8 therefore derives from 427.6. Both sources are cited on this row.

NVIDIA RTX 4000 Ada 20GB

Ada Lovelace PCIe 2023 4nm Spec confidence: Official
FP8 (dense)
214
TFLOPS
FP32
27
TFLOPS
VRAM
20 GB
GDDR6
Bandwidth
360 GB/s
memory
TDP
130 W
205.4 TFLOPS/kW

Overview

The NVIDIA RTX 4000 Ada Generation is a 130 W single-slot professional card with 6,144 CUDA cores, 192 fourth-generation Tensor cores and 48 third-generation RT cores, and 20 GB of GDDR6 with ECC on a 160-bit bus at 360 GB/s. Single slot at 20 GB is an unusual combination and it is why this card turns up in dense workstation and render node builds where four GPUs need to fit in a chassis that will not take dual-slot boards. It delivers 26.7 TFLOPS of FP32 and 213.8 dense TFLOPS of FP8. The same die and the same core count appear in the RTX 4000 SFF Ada Generation, but that board is clocked much lower to fit 70 W, so this is the faster of the two by roughly 40 percent.

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
64-bit floating point
0.4TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
32-bit floating point
27TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
BF16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP8
8-bit floating point
214TFLOPS 428TFLOPS
FP6
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP4
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
INT8
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision

Specifications

Architecture

Ada Lovelace

Form Factor

PCIe

Launch Year

2023

Process Node

4nm

Memory

20 GB GDDR6

Bandwidth

360 GB/s

TDP

130 W

Max Power

130 W

CUDA Cores

6,144

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 6,144
Tensor Cores 192 (4th Gen)
Streaming Multiprocessors 48
Memory
VRAM 20 GB
Memory Type GDDR6
Bandwidth 360 GB/s
Interface Width 160-bit
Interconnect & I/O
PCIe 4.0 x16
Power & Thermal
TDP 130 W
Max Board Power 130 W
Cooling Active
Enterprise Features
ECC Memory Yes
Sparsity Yes
Compute APIs CUDA 12.2, OpenCL 3.0, DirectCompute
Physical & Media
Card Length 241.3 mm
Width 1-slot
NVENC Engines 2
NVDEC Engines 2
General
Form Factor PCIe
Architecture Ada Lovelace
Process Node 4nm
Launch Year 2023

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
No verified data available

Data Quality

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

Rent this GPU

$0.16 / GPU-hr

Lowest of 4 live listings across 3 providers· median $0.24

See all rental prices →

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

Similar GPUs

NVIDIA L40S 48GB

PCIe · 2023
FP32: 92 TFLOPS
Compare vs L40S

NVIDIA L40 48GB

PCIe · 2023
FP32: 91 TFLOPS
Compare vs L40

Frequently Asked Questions

How many TFLOPS does the NVIDIA RTX 4000 Ada have?

The NVIDIA RTX 4000 Ada delivers 27 TFLOPS FP32 and 214 TFLOPS FP8 at peak. Flopper does not currently have a verified FP16 throughput figure for it.

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

The NVIDIA RTX 4000 Ada has a TDP (Thermal Design Power) rating of 130 watts.

How much memory does the NVIDIA RTX 4000 Ada have?

The NVIDIA RTX 4000 Ada is equipped with 20 GB of VRAM with 360 GB/s of memory bandwidth.

What architecture is the NVIDIA RTX 4000 Ada based on?

The NVIDIA RTX 4000 Ada is based on the Ada Lovelace architecture, launched in 2023.

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