NVIDIA Quadro
NVIDIA Professional Visualization Solution
From stunning industrial designs to advanced special effects to complex scientific visualizations, NVIDIA RTX (formerly Quadro) is the ultimate professional visual computing platform. Trusted by creators and technologists to accelerate their workflows, NVIDIA RTX (formerly Quadro) professional GPUs offer a highly advanced ecosystem of hardware, software, and tools to meet today's business challenges. We are transforming it into tomorrow's success.
NVIDIA A40
NVIDIA A40 GPU combines best-in-class professional graphics with powerful computing power to tackle modern design, creative and scientific challenges, powerful AI exploitation, data center performance and multi-workload capabilities has made a revolutionary leap forward. Powering the next generation of virtual workstations and server-based workloads, NVIDIA A40 delivers cutting-edge capabilities for ray-traced rendering, simulation, virtual production and more to professionals anywhere, anytime.
NVIDIA RTX A6000
The NVIDIA RTX A6000 is powered by the NVIDIA Ampere architecture, combining the latest generation of RT Cores, Tensor Cores, and CUDA® Cores with 48GB of graphics memory for extreme desktop performance. It's the world's most powerful graphics solution that delivers the breakthrough innovations designers, engineers and artists need, including unprecedented graphics, rendering and AI performance.
NVIDIA RTX A5000
Powered by the NVIDIA Ampere architecture, the NVIDIA RTX A5000 combines the latest generation of RT Cores, Tensor Cores, and CUDA® Cores with 24GB of graphics memory to deliver the best desktop performance for designers, engineers and artists. increase.
NVIDIA RTX A4000
The NVIDIA RTX A4000 is powered by the NVIDIA Ampere architecture, combining the latest generation of RT Cores, Tensor Cores, and CUDA® Cores with 16GB of graphics memory for extreme desktop performance.
NVIDIA RTX A2000
NVIDIA RTX™ A2000 with NVIDIA RTX technology, real-time ray tracing and AI
Bringing the power of accelerated computing and high-performance graphics to more professionals. Built on the NVIDIA Ampere architecture, the VR-Ready RTX A2000 features 26 2nd Gen RT Cores, 104 3rd Gen Tensor Cores, 3,328 Next Gen CUDA® Cores, and 6GB of Error Correction Code ( Combines GDDR6 graphics memory with ECC). The RTX A2000 features a power-efficient, low-profile dual-slot PCIe form factor that fits in a wide range of small form factor workstations. With the RTX A2000, you can design bigger than ever before, render faster and work smarter.
Product spec
|
NVIDIA RTX A4000 |
NVIDIA RTX A5000 |
NVIDIA RTX A6000 |
NVIDIA A40 |
CUDA Parallel Processing cores
|
6144 |
8192 |
10752 |
|
NVIDIA Tensor Cores
|
192 |
256 |
336 |
|
NVIDIA RT Cores
|
48 |
64 |
84 |
|
Frame buffer memory
|
16GB GDDR6 with ECC |
24GB GDDR6 with ECC |
48GB GDDR6 with ECC |
|
Memory Interface
|
256-bit |
384-bit |
384-bit |
|
Memory Bandwidth
|
448 GB/s |
768 GB/s |
||
Max Power Consumption
|
140W |
230W |
300W |
|
Graphics Bus
|
PCI Express 4.0 x16 |
PCI Express 4.0 x16 |
PCI Express 4.0 x16 |
|
Display Connectors
|
DP 1.4 (4) |
DP 1.4 (4) *1 |
DP 1.4 (4) *2 |
DP 1.4 (3) *3 |
Form factor
|
4.4”H x 9.5”L Single Slot |
4.4”H x 10.5”L Dual Slot |
||
Product Weight
|
500 grams |
1.025kg |
1.179 kg |
987 grams |
Thermal Solution
|
Active |
Passive |
||
Frame lock
|
Compatible (with Quadro Sync II) |
|||
NVLink Interconnect
|
- |
112.5 GB/s (bidirectional) |
112.5 GB/s (bidirectional) |
NVIDIA NVLink 112.5GB/s (bidirectional) PCIe Gen4 16GB/s |
*1 Display ports are on by default for the RTX A5000. Display ports are not active when using vGPU software.
*2 Display ports are on by default for the RTX A6000. Display ports are not active when using vGPU software.
*3 Display ports are not active by default for A40. NVIDIA vGPU software is only supported when display ports are not active.
|
NVIDIA RTX A2000 |
CUDA Parallel Processing cores
|
3328 |
NVIDIA Tensor Cores
|
104 |
NVIDIA RT Cores
|
26 |
Frame buffer memory
|
6GB GDDR6 with ECC |
Memory Interface
|
192-bit |
Memory Bandwidth
|
288 GB/s |
Max Power Consumption
|
70W |
Graphics Bus
|
PCI Express 4.0 x16 |
Display Connectors
|
mDP 1.4a (4) |
Form factor
|
2.7”H x 6.6”L Dual Slot |
Product Weight
|
294g (Low Profile Bracket) 306g (ATX bracket) |
Thermal Solution
|
Active |
Frame lock
|
- |
NVLink Interconnect
|
- |
Quadro RTX 8000
NVIDIA Quadro RTX 8000 opens new frontiers in professional graphics with blazing performance and scalability with 48GB of fast GDDR6 memory and NVIDIA NVLink™. Designers and artists will be able to work beyond what was previously possible with large and complex ray tracing, deep learning and visual computing workloads.
Quadro RTX 6000
Powered by the NVIDIA Turing™ architecture and the NVIDIA RTX platform, the NVIDIA Quadro RTX 6000 has brought the biggest advancement in computer graphics to professional workflows for decades. Designers and artists can take advantage of hardware-accelerated ray tracing, deep learning, and advanced shading to boost productivity and create more amazing content than ever before.
Quadro RTX 5000
Break the boundaries of possibility with NVIDIA Quadro RTX 5000. Powered by the NVIDIA Turing architecture and NVIDIA RTX platform, it blends ray tracing, deep learning, and advanced shading to power next-gen workflows. Creative and technical professionals can make informed decisions faster and easily meet demanding design and visualization workloads.
Product spec
|
Quadro RTX 5000 |
Quadro RTX 6000 |
Quadro RTX 8000 |
CUDA Parallel Processing cores
|
3072 |
4608 |
|
NVIDIA Tensor Cores
|
384 |
576 |
|
NVIDIA RT Cores
|
48 |
72 |
|
Frame buffer memory
|
16GB GDDR6 |
24GB GDDR6 |
48GB GDDR6 with ECC |
RTX-OPS
|
62T |
84T |
|
Rays Cast
|
8 Giga Rays/Sec |
10 Giga Rays/Sec |
|
Peak Single Precision (FP32)
|
11.2 TFLOPS |
16.3 TFLOPS |
|
Performance
|
|||
Peak Half Precision (FP16)
|
22.3 TFLOPS |
32.6 TFLOPS |
|
Performance
|
|||
Peak Integer Operation (INT8)
|
178.4 TOPS |
261.0 TOPS |
|
Performance
|
|||
Deep Learning TeraFLOPS*1
|
89.2 TFLOPS |
130.5 Tensor TFLOPS |
|
Memory Interface
|
256-bit |
384-bit |
384-bit |
Memory Bandwidth
|
448 GB/s |
672 GB/s |
|
Max Power Consumption
|
265W |
295W |
|
Graphics Bus
|
PCI Express 3.0 x16 |
||
Display Connectors
|
DP 1.4 (4) + VirtualLink (1) |
||
Form factor
|
4.4”H x 10.5”L Dual Slot |
||
Product Weight
|
975 grams |
976 grams |
977 grams |
Thermal Solution
|
Active |
||
NVIDIA 3D Vision and 3D Vision Pro
|
Support via 3 pin mini DIN |
||
Frame lock
|
Compatible (with Quadro Sync II) |
||
NVLink Interconnect
|
50GB/s |
100GB/s |
*1 FP16 matrix multiply with FP16 or FP32 accumulate
Quadro RTX for mobile workstations
The lineup is based on a state-of-the-art 12nm FinFET manufacturing process customized for NVIDIA to create the most powerful computing platform for laptop HPC, AI, VR and graphics workloads.
The Turing GPU architecture enabled the biggest leap in computer graphics in over a decade, with 13.6 billion transistors on a single die.
The latest Quadro mobile workstation lineup offers up to 9.4 TFLOPs of peak single-precision computing power and 75.2 TFLOPs of tensor computing power, delivering ultimate graphics and computing performance for the most demanding professional users.
Product spec
|
RTX5000 |
RTX4000 |
RTX3000 |
product image |
|
|
|
CUDA Parallel Processing Cores |
3072 cores |
2560 cores |
2304 cores |
NVIDIA Tensor Cores |
384 |
320 |
288 |
NVIDIA RT Cores |
48 |
40 |
36 |
GPU memory |
16GB GDDR6 Up to 448GB/s |
8GB GDDR6 Up to 448GB/s |
6GB GDDR6 Up to 336GB/s |
Theoretical Peak Single Precision Performance |
Up to 9.4 TFLOPS |
Up to 8.0 TFLOPS |
Up to 6.4 TFLOPS |
Tensor Performance |
Up to 75.2 TOPS |
Up to 63.9 TOPS |
Up to 51.4 TOPS |
Memory Interface |
256-bit |
192-bit |
|
Memory Bandwidth |
Up to 448GB/s |
Up to 336GB/s |
|
Max Power Consumption (TGP) |
80 - 110W |
60-80W |
|
Graphics Bus |
PCI Express 3.0 x16 |
|
T2000 |
T1000 |
P620 |
P520 |
CUDA Parallel Processing Cores |
1024 cores |
768 cores |
512 cores |
384 cores |
GPU memory |
4GB GDDR5 128GB/s |
4GB GDDR5 96GB/s |
2GB GDDR5 48 GB/s |
|
Theoretical Peak Single Precision Performance |
Up to 3.5 TFLOPS |
Up to 2.6 TFLOPS |
Up to 1.5 TFLOPS |
Up to 1.1 TFLOPS |
Memory Interface |
128-bit |
64-bit |
||
Memory Bandwidth |
128GB/s |
96GB/s |
48 GB/s |
|
Max Power Consumption (TGP) |
40-50W |
25W |
18W |
|
Graphics Bus |
PCI Express 3.0 x16 |
* using Advertised Boost clock
Quadro GV100
The new Quadro GV100 has a training capacity of 7.4 TFLOPS double precision, 14.8 TFLOPS single precision and 118.5 TFLOPS, with 32GB of high-bandwidth memory capacity.
Two GV100 cards can be combined using NVIDIA NVLink interconnect technology to expand memory and performance, creating a massive visual computing solution in a single workstation chassis.
Product spec
|
Quadro GV100 |
CUDA parallel processing core |
5120 |
NVIDIA Tensor Cores |
640 |
GPU memory |
32GB HBM2 |
FP64 performance |
7.4 TFLOPS |
FP32 Performance |
14.8 TFLOPS |
FP16 Performance |
29.6 TFLOPS |
Tensor performance |
118.5 TFLOPS |
Maximum power consumption |
250W |
graphics bus |
PCI Express 3.0 x16 |
display connector |
DP 1.4 (4) |
form factor |
4.4”H x 10.5”L dual slot |
NVLink Bridge for GV100
2 NVLink Bridge for GV100 allows two GV100s to communicate faster than traditional PCIe bus communication. GV100 adopts the 2nd generation NVLink technology, delivering 100GB/s (bi-directional) GPU-to-GPU data transfer speed.
Quadro P2200
With the perfect balance of performance, compelling features, and compact form factor, the Quadro P2200 offers great creative experience and productivity in a variety of professional 3D applications.
It features a Pascal GPU with 1280 CUDA cores, 5 GB of massive GDDR5X onboard memory, and power to drive up to four 5K (5120 x 2880 @ 60 Hz) displays natively.
Accelerate your product development and content creation workflows with GPUs that provide the fluid interactivity you need to work with large scenes and models.
Product spec
|
Quadro P2200 |
NVIDIA CUDA® Parallel Processing Cores |
1280 |
Frame buffer memory |
5GB GDDR5x |
Peak Single-Precision Performance |
Up to 3.8 TFLOPS |
Memory Interface |
160-bit |
Memory Bandwidth |
Up to 200GB/s |
Max Power Consumption |
75W |
Graphics Bus4 |
PCI Express 3.0 x16 |
Display Connectors |
DP 1.4 (4) |
Form factor |
4.4”H x 7.9”L Single Slot |
Product Weight |
256 grams |
Thermal Solution |
Active |
Other Quadro products
Product spec
|
Quadro GP100 |
Quadro P6000 |
Quadro P5000 |
product image |
|
|
|
CUDA parallel processing core |
3584 |
3840 |
2560 |
GPU memory |
16GB HBM2 |
24GB GDDR5X |
16GB GDDR5X |
FP64 performance |
5.2 TFLOPS |
- |
- |
FP32Performance |
10.3 TFLOPS |
12 TFLOPS |
8.9 TFLOPS |
FP16 Performance |
20.7 TFLOPS |
- |
- |
Maximum power consumption |
235W |
250W |
180W |
graphic bus |
PCI Express 3.0 x16 |
||
display connector |
DP 1.4 (4), DVI-D (1), Optional Stereo (1) |
||
form factor |
4.4”H x 10.5”L Dual Slot |
|
Quadro P4000 |
Quadro P2000 |
Quadro P1000 |
product image |
|
|
|
CUDA parallel processing core |
1792 |
1024 |
640 |
GPU memory |
8GB GDDR5 |
5GB GDDR5 |
4GB GDDR5 |
FP32Performance |
5.3 TFLOPS |
3.0 TFLOPS |
1.894 TFLOPS |
Maximum power consumption |
105W |
75W |
47W |
graphic bus |
PCI Express 3.0 x16 |
||
display connector |
DP 1.4 (4), Optional Stereo (1) |
DP 1.4 (4) |
mDP 1.4 (4) |
form factor |
4.4”H x 9.5”L Single Slot |
2.713”H x 5.906”L, Single Slot, Low Profile |
|
Quadro Synk II |
product image |
|
Per Sync II |
Up to 16 Displays Up to 4 GPUs |
Per system with 2 Sync II |
Up to 32 Displays Up to 8 GPUs |
per cluster (1 or 2 Sync II per Node) |
Up to 50 Nodes Up to 200 GPUs |
House Sync |
TTL/Composite Input or TTL Output |
Power Connectors |
6-pin PCI or SATA |
Form factor |
4.2”H x 6”L Single Slot |
|
Quadro NVLink Bridge |
product image
|
|
Bandwidth
|
Up to 80 GB/s bi-directional (with two bridges) |
Form factor
|
2-Way 2-Slot Spacing |
|
Quadro SLI HB Bridge QUICK SPECS |
product image |
|
Resolution Supported |
Up to 4K @ 60Hz Up to 5K @ 60Hz |
Form factor |
2-Way 2-Slot Spacing and 2-Way 3-Slot Spacing |