NVIDIA · GeForce RTX 40 Series

NVIDIA GeForce RTX 4090 D

The NVIDIA GeForce RTX 4090 D is a graphics card released on December 28, 2023, built on the Ada Lovelace architecture. It has 14,592 shading units, 24 GB of GDDR6X memory on a 384-bit bus, 1,008.0 GB/s of memory bandwidth, a boost clock of 2.52 GHz and a 425 W board power rating. It scores 30,429 in the PassMark G3D benchmark. It supports OpenGL 4.6, Vulkan 1.4 and CUDA compute capability 8.9.

Ada LovelaceAD102-250-A1

10/10 key specs recorded · 33 facts

Depth: 34 recorded fields, 3 features, 5 PassMark workloads

Specification data last verified October 1, 2026 against PassMark Videocard Benchmark and Wikipedia GPU list tables (wikigrid-2026-09). Coverage: 10 of 10 tracked specifications recorded.

PassMark Videocard Benchmark Wikipedia GPU list tables (wikigrid-2026-09)PassMark Videocard Benchmark

Shading units

14,592cores

114 SMs

Memory

24 GB

GDDR6X · 384-bit

Boost clock

2.52 GHz

2.28 GHz base

Board power

425 W

Performance

The NVIDIA GeForce RTX 4090 D scores 30,429 in the PassMark G3D benchmark, higher than 99.3% of the 2,658 GPUs benchmarked (PassMark rank #21), measured across 34 submitted results. Its PassMark G2D score is 985.

PassMark G3D Mark

30,429

Rank #21
Faster than 99.3% of 2,658 GPUs
34 samples

0relative to the fastest GPU on record (41,928)

PassMark workload tests

DirectX 9 343 fps

reference: fastest recorded 526 fps · faster than 98.4% of 1,675 GPUs

DirectX 10 178 fps

reference: fastest recorded 231 fps · faster than 98.3% of 1,602 GPUs

DirectX 11 286 fps

reference: fastest recorded 362 fps · faster than 98.5% of 1,430 GPUs

DirectX 12 114 fps

reference: fastest recorded 213 fps · faster than 98% of 1,274 GPUs

GPU Compute 19,000 ops/s

reference: fastest recorded 26,591 ops/s · faster than 98.9% of 1,447 GPUs

Averages of PassMark PerformanceTest submissions; frames/sec are PassMark test-scene rates, not game FPS. Tests PassMark does not report for this card are omitted. PassMark listing

In the PassMark workload tests the NVIDIA GeForce RTX 4090 D averages DirectX 9 343 fps, DirectX 10 178 fps, DirectX 11 286 fps, DirectX 12 114 fps and GPU Compute 19,000 ops/s, across 34 submitted results. Frames per second are PassMark test-scene rates, not game frame rates.

Theoretical throughput

FP32 (single precision) 73.544 TFLOPS calculatedcalculated from shaders x boost clock (MHz) x 2 / 1e6 = 14592 x 2520 x 2 / 1e6

FP32 (single precision) (73.544 TFLOPS) is calculated by Feex from stored specifications, not stated by NVIDIA.

As stated by PassMark

First benchmarked 2024-03-05

Samples 34

PassMark listing (collected 2026-10-01)

Not to be confused with

This page covers the NVIDIA GeForce RTX 4090 D specifically. It is a different product from NVIDIA GeForce RTX 4090, NVIDIA GeForce RTX 4090 Laptop GPU, which have their own specifications. Check the exact model name before comparing figures.

Graphics processor

Architecture Ada Lovelace

Codename AD102-250-A1

Chip Not recorded

Process node Not recorded

Transistors 76,300 M

Die size 609 mm²

The NVIDIA GeForce RTX 4090 D is based on the Ada Lovelace architecture, uses the AD102-250-A1 graphics processor, contains 76,300 million transistors and has a die size of 609 mm².

Shader configuration

Shading units 14,592

Compute units Not recorded

Streaming multiprocessors 114

Execution units Not recorded

Texture units (TMUs) 456

Render output units (ROPs) 176

Ray tracing cores 114

Tensor / AI cores 456

XMX engines Not recorded

Infinity cache Not recorded

The NVIDIA GeForce RTX 4090 D has 14,592 shading units, 114 streaming multiprocessors, 456 texture mapping units, 176 render output units, 114 ray tracing cores and 456 tensor cores.

The NVIDIA GeForce RTX 4090 D does not use XMX engines; those are specific to Intel Arc graphics. Infinity Cache is an AMD architectural feature and does not apply to the NVIDIA GeForce RTX 4090 D.

Clock speeds

Base clock 2.28 GHz

Boost clock 2.52 GHz

Game clock Not recorded

Memory clock Not recorded

Effective memory clock Not recorded

The NVIDIA GeForce RTX 4090 D runs at a base clock of 2.28 GHz and boosts to 2.52 GHz.

Memory

Size 24 GB

Type GDDR6X

Bus width 384-bit

Bandwidth 1,008.0 GB/s

The NVIDIA GeForce RTX 4090 D is equipped with 24 GB of GDDR6X memory, uses a 384-bit memory bus and delivers 1,008.0 GB/s of memory bandwidth.

Board and power

The NVIDIA GeForce RTX 4090 D has a board power rating of 425 W.

Board power (TDP) 425 W

Total board power Not recorded

Suggested PSU Not recorded

Power connectors Not recorded

Slot width Not recorded

Length Not recorded

Width Not recorded

Height Not recorded

Cooler Not recorded

Max temperature Not recorded

Bus interface Not recorded

Market

Launched December 28, 2023

Launch price Not recorded

Status Not recorded

Form factor Not recorded

Segment Desktop

Series GeForce RTX 40 Series

The NVIDIA GeForce RTX 4090 D was released on December 28, 2023.

Benchmarks

PassMark G3D 30,429

PassMark G2D 985

Rank #21

Samples 34

Graphics and compute APIs

The NVIDIA GeForce RTX 4090 D supports OpenGL 4.6, Vulkan 1.4 and CUDA compute capability 8.9.

OpenGL 4.6

Vulkan 1.4

CUDA compute capability 8.9

Feex has no recorded OpenCL version for the NVIDIA GeForce RTX 4090 D.

Features

The NVIDIA GeForce RTX 4090 D has 3 supported features recorded outside the media engine. A feature that no source mentions is shown as not listed, never as unsupported.

Rendering

Supported: Hardware ray tracing 3rd Generation

Dedicated GPU hardware that speeds up ray-traced lighting, reflections and shadows in supported games and apps.

Ray tracing units: 114

Not listed (3)

No source Feex holds states whether the NVIDIA GeForce RTX 4090 D supports DirectX 12 Ultimate, NVIDIA DLSS and Vulkan ray tracing. That is silence, not a statement that they are unsupported.

The NVIDIA GeForce RTX 4090 D supports Hardware ray tracing (3rd Generation) under Rendering.

Compute and AI

Supported: CUDA Compute capability 8.9

Support for NVIDIA's CUDA computing platform, used by many creative, scientific and AI applications.
Supported: Tensor cores 4th Generation

NVIDIA's dedicated matrix-math units used for AI workloads such as DLSS and machine-learning inference.

Tensor cores: 456

Not listed (1)

No source Feex holds states whether the NVIDIA GeForce RTX 4090 D supports NVLink / SLI. That is silence, not a statement that they are unsupported.

The NVIDIA GeForce RTX 4090 D supports CUDA (Compute capability 8.9) and Tensor cores (4th Generation) under Compute and AI.

Frequently asked questions

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

The NVIDIA GeForce RTX 4090 D has 24 GB of GDDR6X memory on a 384-bit bus. Its memory bandwidth is 1,008.0 GB/s.

How much power does the NVIDIA GeForce RTX 4090 D use and what power supply does it need?

The NVIDIA GeForce RTX 4090 D has a board power rating of 425 W.

How fast is the NVIDIA GeForce RTX 4090 D?

The NVIDIA GeForce RTX 4090 D scores 30,429 in the PassMark G3D benchmark, higher than 99.3% of the 2,658 GPUs benchmarked. Its PassMark rank is #21. In the PassMark DirectX 12 test it averages 114 fps.

Does the NVIDIA GeForce RTX 4090 D support ray tracing?

Yes. The NVIDIA GeForce RTX 4090 D has hardware ray tracing (3rd Generation) with 114 ray tracing cores.

What is the CUDA compute capability of the NVIDIA GeForce RTX 4090 D?

The NVIDIA GeForce RTX 4090 D has CUDA compute capability 8.9.

When was the NVIDIA GeForce RTX 4090 D released?

The NVIDIA GeForce RTX 4090 D was released on December 28, 2023.

Corrections to this record

After an audit, 1 value on the NVIDIA GeForce RTX 4090 D record was corrected. Each change is listed with its old value, new value, rule and sources.

Launch price: ~~12,999 USD~~ cleared (now Not recorded). Rule: Implausible launch price removed. above the $3,000 consumer ceiling (not a USD launch price).

The audit log is public: all changes for this GPU, the corrections overview and the method.