NVIDIA · RTX PRO

NVIDIA RTX PRO 4000 Blackwell

The NVIDIA RTX PRO 4000 Blackwell is a graphics card released on March 18, 2025, built on the Blackwell architecture. It has 8,960 shading units and a boost clock of 2.62 GHz. It scores 28,235 in the PassMark G3D benchmark. It supports Vulkan 1.3, OpenCL 3.0 and CUDA compute capability 12.0.

BlackwellGB203

6/10 key specs recorded · 38 facts

Depth: 26 recorded fields, 10 features, 5 PassMark workloads

Specification data last verified October 1, 2026 against PassMark Videocard Benchmark, NVIDIA Developer: CUDA GPUs, NVIDIA Developer: Video Encode and Decode GPU Support Matrix and Wikipedia GPU list tables (wikigrid-2026-09). Coverage: 6 of 10 tracked specifications recorded.

PassMark Videocard Benchmark NVIDIA Developer: CUDA GPUs NVIDIA Developer: Video Encode and Decode GPU Support Matrix Wikipedia GPU list tables (wikigrid-2026-09)PassMark Videocard Benchmark

Shading units

8,960cores

Memory

n/a

GDDR7

Boost clock

2.62 GHz

1.59 GHz base

Board power

n/a

Performance

The NVIDIA RTX PRO 4000 Blackwell scores 28,235 in the PassMark G3D benchmark, higher than 99% of the 2,658 GPUs benchmarked (PassMark rank #32), measured across 247 submitted results. Its PassMark G2D score is 1,261.

PassMark G3D Mark

28,235

Rank #32
Faster than 99% of 2,658 GPUs
247 samples

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

PassMark workload tests

DirectX 9 355 fps

reference: fastest recorded 526 fps · faster than 98.6% of 1,675 GPUs

DirectX 10 174 fps

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

DirectX 11 269 fps

reference: fastest recorded 362 fps · faster than 97.7% of 1,430 GPUs

DirectX 12 94 fps

reference: fastest recorded 213 fps · faster than 95.6% of 1,274 GPUs

GPU Compute 14,519 ops/s

reference: fastest recorded 26,591 ops/s · faster than 97.2% 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 RTX PRO 4000 Blackwell averages DirectX 9 355 fps, DirectX 10 174 fps, DirectX 11 269 fps, DirectX 12 94 fps and GPU Compute 14,519 ops/s, across 247 submitted results. Frames per second are PassMark test-scene rates, not game frame rates.

As stated by PassMark

Category Desktop

First benchmarked 2025-09-17

Samples 247

PassMark listing (collected 2026-10-01)

Not to be confused with

This page covers the NVIDIA RTX PRO 4000 Blackwell specifically. It is a different product from NVIDIA RTX PRO 4000 Blackwell Generation Laptop GPU, NVIDIA RTX PRO 4000 Blackwell SFF Edition, NVIDIA RTX PRO 4000 Blackwell Mobile, which have their own specifications. Check the exact model name before comparing figures.

Graphics processor

Architecture Blackwell

Codename GB203

Chip Not recorded

Process node Not recorded

Transistors Not recorded

Die size Not recorded

The NVIDIA RTX PRO 4000 Blackwell is based on the Blackwell architecture and uses the GB203 graphics processor.

Shader configuration

Shading units 8,960

Compute units Not recorded

Streaming multiprocessors Not recorded

Execution units Not recorded

Texture units (TMUs) 280

Render output units (ROPs) 96

Ray tracing cores Not recorded

Tensor / AI cores Not recorded

XMX engines Not recorded

Infinity cache Not recorded

The NVIDIA RTX PRO 4000 Blackwell has 8,960 shading units, 280 texture mapping units and 96 render output units.

The NVIDIA RTX PRO 4000 Blackwell 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 RTX PRO 4000 Blackwell. Feex has no recorded ray tracing core count for the NVIDIA RTX PRO 4000 Blackwell.

Clock speeds

Base clock 1.59 GHz

Boost clock 2.62 GHz

Game clock Not recorded

Memory clock Not recorded

Effective memory clock Not recorded

The NVIDIA RTX PRO 4000 Blackwell runs at a base clock of 1.59 GHz and boosts to 2.62 GHz.

Memory

Size Not recorded

Type GDDR7

Bus width Not recorded

Bandwidth Not recorded

Market

Launched March 18, 2025

Launch price $1,546

Status Not recorded

Form factor Not recorded

Segment Workstation

Series RTX PRO

The NVIDIA RTX PRO 4000 Blackwell was released on March 18, 2025 and launched at a price of $1,546.

Benchmarks

PassMark G3D 28,235

PassMark G2D 1,261

Rank #32

Samples 247

Graphics and compute APIs

The NVIDIA RTX PRO 4000 Blackwell supports Vulkan 1.3, OpenCL 3.0 and CUDA compute capability 12.0.

Vulkan 1.3

OpenCL 3.0

CUDA compute capability 12.0

Features

The NVIDIA RTX PRO 4000 Blackwell has 1 supported feature recorded outside the media engine. A feature that no source mentions is shown as not listed, never as unsupported.

Compute and AI

Supported: CUDA Compute capability 12.0

Support for NVIDIA's CUDA computing platform, used by many creative, scientific and AI applications.

Not listed (2)

No source Feex holds states whether the NVIDIA RTX PRO 4000 Blackwell supports Tensor cores and NVLink / SLI. That is silence, not a statement that they are unsupported.

The NVIDIA RTX PRO 4000 Blackwell supports CUDA (Compute capability 12.0) under Compute and AI.

Media engine

The NVIDIA RTX PRO 4000 Blackwell provides hardware encode for H.264 (AVC), HEVC (H.265) and AV1 and hardware decode for H.264 (AVC), HEVC (H.265), AV1 and VP9.

NVIDIA NVENC encoder 9th Generation Encoder engines: 2NVIDIA's dedicated video-encoding engine, used by streaming and capture software; newer generations encode more efficiently.

NVIDIA NVDEC decoder 6th Generation Decoder engines: 2NVIDIA's dedicated video-decoding engine; newer generations support more codecs and formats.

Hardware video codec support of the NVIDIA RTX PRO 4000 Blackwell
Codec	Encode	Decode
H.264 (AVC)	Supported4:4:4: Yes	Supported
HEVC (H.265)	Supported4:2:2: Yes; 4:4:4: Yes; 10-bit: Yes; B-frames: Yes; 8K: Yes	Supported4:4:4: Yes
AV1	Supported	Supported
VP9	Not stated by the vendor	Supported

supported explicitly not supported not stated by the vendor

Media engines recorded for the NVIDIA RTX PRO 4000 Blackwell: NVIDIA NVENC encoder (9th Generation, encoder engines: 2) and NVIDIA NVDEC decoder (6th Generation, decoder engines: 2).

Frequently asked questions

How fast is the NVIDIA RTX PRO 4000 Blackwell?

The NVIDIA RTX PRO 4000 Blackwell scores 28,235 in the PassMark G3D benchmark, higher than 99% of the 2,658 GPUs benchmarked. Its PassMark rank is #32. In the PassMark DirectX 12 test it averages 94 fps.

Does the NVIDIA RTX PRO 4000 Blackwell support AV1 encoding?

Yes. The NVIDIA RTX PRO 4000 Blackwell supports AV1 hardware encoding and AV1 hardware decoding.

What is the CUDA compute capability of the NVIDIA RTX PRO 4000 Blackwell?

The NVIDIA RTX PRO 4000 Blackwell has CUDA compute capability 12.0.

When was the NVIDIA RTX PRO 4000 Blackwell released and how much did it cost at launch?

The NVIDIA RTX PRO 4000 Blackwell was released on March 18, 2025 and launched at a price of $1,546.

Corrections to this record

After an audit, 1 empty field was filled from sources. Each change is listed with its old value, new value, rule and sources.

1 filled value

Architecture: Blackwell (was Not recorded). Rule: Architecture from chip codename. chip GB203 -> Blackwell. Sources: Wikipedia: List of Nvidia graphics processing units (revision 1377107724).

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