RTX 4000 Ada Generation

Graphics processor

Video memory: 20 GB^*
Memory bus: 160 bit^*
Rated memory bandwidth: 360 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Ada Lovelace^*
Process node: 5 nm^*
GPU base clock: 1,500 MHz^*
GPU boost clock: 2,175 MHz^*

Memory & storage

Video memory: 20 GB^*
Video memory type: Reported value 1: GDDR6^*
Reported value 2: GDDR6X^*
Memory bus: 160 bit^*
Rated memory bandwidth: 360 GB/s^*

System configuration

Detailed specifications

Specifications 1 · 15 fields

GPU architecture — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

Ada Lovelace^*

Theoretical peak: FP32 non-tensor dense per GPU — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

26.7^* TFLOPS

Native hardware support: FP4 tensor — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

No^*

Native hardware support: FP6 tensor — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

No^*

Calculated theoretical peak: FP8 tensor dense per GPU — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

163.8^* TFLOPS

Scope and conditions

Calculation: 327.6 / 2^*

Theoretical peak: FP8 tensor with sparsity per GPU — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

327.6^* TFLOPS

Physical configuration — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

PCIe, full height, single slot (4.4 in H x 9.5 in L)^*

Native hardware support for GPU interconnect — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

No^*

PCIe interface declaration — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

PCIe Gen 4 x16^*

Maximum board power — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

130^* W

Peak memory bandwidth per GPU — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

360^* GB/s

Memory interface width per GPU — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

160^* bit

Memory capacity per GPU — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

20^* GB

Memory technology — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

GDDR6^*

GPU model — NVIDIA RTX 4000 Ada Generation — 20 GB GDDR6

NVIDIA RTX 4000 Ada Generation^*

Specifications 2 · 8 fields

Product Name: RTX 4000 Ada Generation^*
GPU Chip: AD104^*
Released: Aug 9th, 2023^*
Bus: PCIe 4.0 x16^*
Memory: 20 GB, GDDR6, 160 bit^*
GPU clock: 1,500^* MHz
Memory clock: 1,750^* MHz
Shaders / TMUs / ROPs: 6144 / 192 / 80^*

Specifications 3 · 57 fields

Brand: NVIDIA^*
Top BUS WIDTH: 160^* bit
Graphics Processor GPU Name: AD104^*
Graphics Processor Architecture: Ada Lovelace^*
Render Config SM Count: 48^*
Graphics Features CUDA: 8.9^*
Render Config Tensor Cores: 192^*
Graphics Processor Process Size: 5^* nm
Graphics Processor Transistors: 35,800 million^*
Graphics Processor Die Size: 294^* mm²
Graphics Card Release Date: Aug 9th, 2023^*
Graphics Card Generation: Workstation Ada (x000A)^*
Graphics Card Successor: Blackwell PRO W^*
Graphics Card Production: Active^*
Name: RTX 4000 Ada Generation^*
Graphics Card Bus Interface: PCIe 4.0 x16^*
Clock Speeds Memory Clock: 2250 MHz 18 Gbps effective^*
Memory Memory Size: 20^* GB
Memory Memory Type: GDDR6^*
Memory Memory Bus: 160^* bit
Memory Bandwidth: 360^* GB/s
Render Config TMUs: 192^*
Top GRAPHICS PROCESSOR: AD104^*
Render Config ROPs: 64^*
Theoretical Performance Pixel Rate: 139.2 GPixel/s
Theoretical Performance Texture Rate: 417.6 GTexel/s
Board Design Slot Width: Single-slot^*
Board Design TDP: 130^* W
Board Design Suggested PSU: 300^* W
Board Design Outputs: 4x DisplayPort 1.4a^*
Graphics Features DirectX: 12 Ultimate (12_2)^*
Graphics Features OpenGL: 4.6^*
Graphics Features OpenCL: 3^*
Graphics Features Vulkan: 1.3^*
source file: nvidia2023.json^*
Graphics Processor Foundry: TSMC^*
Graphics Card Predecessor: Workstation Ampere^*
Recommended Resolutions: Good:1920x1080 | Good:2560x1440 | Playable:3840x2160^*
Board Design Length: 245 mm 9.6 inches^*
Board Design Power Connectors: 1x 16-pin^*
Graphics Processor Density: 121.8M / mm²^*
Board Design Height: 112 mm 4.4 inches^*
Top TMUS: 192^*
Top ROPS: 64^*
Clock Speeds Base Clock: 1,500^* MHz
Clock Speeds Boost Clock: 2,175^* MHz
Graphics Features Shader Model: 6.8^*
Top CORES: 6,144^*
Render Config Shading Units: 6,144^*
Render Config L2 Cache: 48^* MB
Top MEMORY SIZE: 20^* GB
Theoretical Performance FP32 (float): 26.73^* TFLOPS
Theoretical Performance FP64 (double): 417.6 GFLOPS (1:64)
Render Config L1 Cache: 128 KB (per SM)^*
Theoretical Performance FP16 (half): 26.73 TFLOPS (1:1)
Top MEMORY TYPE: GDDR6^*
Render Config RT Cores: 48^*

Specifications 4 · 26 fields

model: RTX 4000 Ada Generation^*
base clock mhz: 1,500^* MHz
max boost clock mhz: 2,175^* MHz
bus type: GDDR6^*
bus width: 160^*
vram: 20^*
clock speed: 14,000^*
bandwith: 360^*
fp16: 26,730^*
fp32: 26,730^*
fp64: 417^*
price: 1,878.95^*
tdp: 130^*
nv link: No^*
bus interface: PCIe 4.0 x16^*
profile: FHFL^*
slot: Single^*
eol: No^*
eol date: No^*
market: Professional^*
ecc equipped: Yes^*
architecture: Ada Lovelace^*
release date: 8/9/2023^*
tensor cores: 192^*
tensor core gen: 4th Gen (Ada)^*
cuda cores: 6,144^*

Specifications 5 · 14 fields

manufacturer: NVIDIA^*
rop: 80^*
igp: No^*
bus: PCIe 4.0 x16^*
memType: GDDR6X^*
gpuChip: GA104^*
productName: RTX 4000 Ada Generation^*
releaseYear: 2,023^*
memSize: 20^*
memBusWidth: 160^*
gpuClock: 1,410^*
memClock: 1,188^*
unifiedShader: 4,864^*
tmu: 152^*

* Provisional data