China only. Unlike the RTX 5090 D, this card is not throttled inside the Tensor Core: it is simply a smaller AD102 bin, 14,592 cores against 16,384, and its published 1,177 AI TOPS is exactly sixteen times its FP32 rate, the same ratio every other Ada GeForce card shows. The whole precision ladder therefore applies normally.

NVIDIA GeForce RTX 4090 D 24GB

Ada Lovelace PCIe 2023 4nm Spec confidence: Official
FP8 (dense)
294
TFLOPS
FP32
74
TFLOPS
VRAM
24 GB
GDDR6X
TDP
425 W
173.0 TFLOPS/kW

Overview

The GeForce RTX 4090 D was NVIDIA's China export version of the RTX 4090, announced at the end of December 2023 to stay inside United States export limits. It cuts the AD102 die from 128 SMs to 114, giving 14,592 CUDA cores and 456 fourth-generation Tensor cores, and drops board power from 450 W to 425 W, while keeping the RTX 4090's 24 GB of GDDR6X on a 384-bit bus and its 2.52 GHz boost clock. That works out to 73.5 TFLOPS of FP32 against the 4090's 82.6, about eleven percent less, and NVIDIA's own 1,177 AI TOPS figure confirms the Tensor cores scale with it rather than being separately capped. It appears regularly in Chinese rental supply and is often listed simply as a 4090, so the core count is worth checking.

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
64-bit floating point
1.1TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
32-bit floating point
74TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
TensorFloat-32
74TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
BF16
Brain Float 16
147TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP16
16-bit floating point
147TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP8
8-bit floating point
294TFLOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
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
8-bit integer
588TOPS The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision

The GeForce RTX 4090 D 24GB 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 MI355XGeForce RTX 4090 D 24GB

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

Specifications

Architecture

Ada Lovelace

Form Factor

PCIe

Launch Year

2023

Process Node

4nm

Memory

24 GB GDDR6X

Bandwidth

No verified data available

TDP

425 W

Max Power

425 W

Transistors

76.3 billion

CUDA Cores

14,592

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 14,592
Tensor Cores 456 (4th Gen)
Streaming Multiprocessors 114
Base Clock 2.28 GHz
Boost Clock 2.52 GHz
Chip Design
Transistors 76.3 billion
Die Size 609 mm²
Process Node 4nm
Memory
VRAM 24 GB
Memory Type GDDR6X
Bandwidth No verified data available
Interface Width 384-bit
Interconnect & I/O
PCIe 4.0
Power & Thermal
TDP 425 W
Max Board Power 425 W
Cooling Active
Enterprise Features
Sparsity Yes
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
2023-12-28

Data Quality

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

Rent this GPU

$0.40 / 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

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 GeForce RTX 4090 D have?

The NVIDIA GeForce RTX 4090 D delivers 74 TFLOPS FP32, 147 TFLOPS FP16 and 294 TFLOPS FP8 at peak.

What is the power consumption of the NVIDIA GeForce RTX 4090 D?

The NVIDIA GeForce RTX 4090 D has a TDP (Thermal Design Power) rating of 425 watts.

How much memory does the NVIDIA GeForce RTX 4090 D have?

The NVIDIA GeForce RTX 4090 D is equipped with 24 GB of VRAM. Flopper does not currently have a verified memory bandwidth figure for it.

What architecture is the NVIDIA GeForce RTX 4090 D based on?

The NVIDIA GeForce RTX 4090 D 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