This is custom heading element
The Heart of the Modern Data Center
NVIDIA Tesla GPU’s Types
NVIDIA A100 TENSOR CORE GPU
The Most Advanced Data Center GPU Ever Built.
NVIDIA® V100 is the world’s most advanced data center GPU ever built to accelerate AI, HPC, and graphics. Powered by NVIDIA Volta, the latest GPU architecture, V100 offers the performance of up to 100 CPUs in a single GPU—enabling data scientists, researchers, and engineers to tackle challenges that were once thought impossible.
SPECIFICATIONS
| System Interface | NVIDIA NVLink | NVIDIA PCIe |
| GPU Memory | 40GB – 80GB | 40 GB |
| GPU Memory Bandwidth | 1,555 GB/s – 2,039 GB/s | 1,555 GB/s |
| Peak FP64 | 9.7 TF | |
| Peak FP64 Tensor Core | 19.5 TF | |
| Peak FP32 | 19.5 TF | |
| Peak FP32 Tensor Core | 156 TF – 312 TF* | |
| Peak BFLOAT16 Tensor Core | 312 TF – 624 TF* | |
| Peak FP16 Tensor Core | 312 TF – 624 TF* | |
| Peak INT8 Tensor Core | 624 TOPS – 1,248 TOPS* | |
| Peak INT4 Tensor Core | 1,248 TOPS – 2,496 TOPS* | 1,248 TOPS |
| Interconnect | NVIDIA NVLink 600 GB/s PCIe Gen4 64 GB/s | |
| Multi-Instance GPU | Various instance sizes with up to 7 MIGs @ 10 GB | Various instance sizes with up to 7 MIGs @ 5 GB |
| Form Factor | 4/8 SXM on NVIDIA HGX™ A100 | PCIe |
| Max TDP Power | 400 W | 250 W |
| Delivered Performance of Top Apps | 100% | 90% |
NVIDIA V100 PCIe
The Most Advanced Data Center GPU Ever Built.
NVIDIA® V100 is the world’s most advanced data center GPU ever built to accelerate AI, HPC, and graphics. Powered by NVIDIA Volta, the latest GPU architecture, V100 offers the performance of up to 100 CPUs in a single GPU—enabling data scientists, researchers, and engineers to tackle challenges that were once thought impossible.
SPECIFICATIONS
| GPU Architecture | NVIDIA Volta | |
| NVIDIA Tensor Cores | 640 | |
| NVIDIA CUDA® Core | 5,120 | |
| Double-Precision Performance | 7 TFLOPS | 7.8 TFLOPS |
| Single-Precision Performance | 14 TFLOPS | 15.7 TFLOPS |
| Tensor Performance | 112 TFLOPS | 125 TFLOPS |
| GPU Memory | 32GB /16GB HBM2 | |
| Memory Bandwidth | 900GB/sec | |
| ECC | Yes | |
| Interconnect Bandwidth | 32GB/sec | 300GB/sec |
| System Interface | PCIe Gen3 | NVIDIA NVLink |
| Form Factor | CIe Full Height/Length | SXM2 |
| Max Power Consumption | 250 W | 300 W |
| Thermal Solution | Passive | |
| Compute APIs | CUDA, DirectCompute, OpenCL™, OpenACC | |
NVIDIA T4 - TENSOR CORE GPU
GPU Acceleration Goes Mainstream
NVIDIA T4 enterprise GPUs fits into the most mainstream servers and standard data center infrastructures. It has NVIDIA Turing™ Tensor Cores architecture, low-profile, 70-watt (W) and delivers revolutionary multi-precision performance to accelerate a wide range of applications, machine learning, deep learning, and virtual desktops.
SPECIFICATIONS
| GPU Architecture | NVIDIA Turing | |
| NVIDIA Turing Tensor Cores | 320 | |
| NVIDIA CUDA® Core | 2,560 | |
| Single-Precision | 8.1 TFLOPS | |
| Mixed-Precision (FP16/FP32) | 65 TFLOPS | |
| INT8 | 130 TOPS | |
| INT4 | 260 TOPS | |
| GPU Memory | 16 GB GDDR6 300 GB/sec |
|
| ECC | Yes | |
| Interconnect Bandwidth | 32 GB/sec | |
| System Interface | x16 PCIe Gen3 | |
| Form Factor | Low-Profile PCIe | |
| Thermal Solution | Passive | |
| Compute APIs | CUDA, NVIDIA TensorRT™, ONNX | |
NVIDIA Quadro RTX™ 6000
QUADRO POWERED SERVERS
The RTX 6000 is optimized for reliability in enterprise data centers and built for 24/7 server environments. It features a passive thermal solution to fit into a variety of servers. Tackle graphics-intensive workloads such as batch rendering, virtualization, data science, simulation, and scientific visualization , all powered by NVIDIA RTX.
SPECIFICATIONS
| GPU memory | 24 GB GDDR6 | |
| Memory interface | 384-bit | |
| Memory Bandwidth | Up to 624 GB/s | |
| Error-correcting code (ECC) | Yes | |
| NVIDIA CUDA Cores | 4,608 | |
| NVIDIA Tensor Cores | 576 | |
| NVIDIA RT Cores | 72 | |
| Single-Precision Performance | 14.9 TFLOPS | |
| Tensor Performance | 119.4 TFLOPS | |
| NVIDIA NVLink | Yes | |
| NVIDIA NVLink bandwidth | 100 GB/s (bidirectional) | |
| System Interface | PCI Express 3.0 x 16 | |
| Power Consumption | 250W | |
| Thermal Solution | Passive | |
| Form Factor | 4.4” H x 10.5” L dual slot | |
| Encode/decode engines | Encode/decode engines | |
| Display connectors | None³ | |
| NVIDIA Driver Requirement | R440 U2 and later | |
| Graphics APIs | Shader Model 5.1, OpenGL 4.5, DirectX 12 | |
| Compute APIs | CUDA, DirectCompute, OpenCL™,OpenACC® | |
NVIDIA® Quadro RTX™ 8000
REAL TIME RAY TRACING FOR PROFESSIONALS
Experience unbeatable performance, power, and memory with the NVIDIA® Quadro RTX™ 8000, the world’s most powerful graphics card for professional workflows.
SPECIFICATIONS
| GPU memory | 48 GB GDDR6 | |
| Memory interface | 384-bit | |
| Memory Bandwidth | 672 GB/s | |
| Error-correcting code (ECC) | Yes | |
| NVIDIA CUDA Cores | 4,608 | |
| NVIDIA Tensor Cores | 576 | |
| NVIDIA RT Cores | 72 | |
| Single-Precision Performance | 16.3 TFLOPS | |
| Tensor Performance | 130.5 TFLOPS | |
| NVIDIA NVLink | Connects 2 Quadro RTX 8000 GPUs1 | |
| NVIDIA NVLink bandwidth | 100 GB/s (bidirectional) | |
| System Interface | PCI Express 3.0 x 16 | |
| Power Consumption | Total board power: 295 W / Total graphics power: 260 W | |
| Thermal Solution | Active | |
| Form Factor | 4.4” H x 10.5” L, Dual Slot, Full Height | |
| Encode/decode engines | 1X Encode, 1X Decode | |
| Display connectors | 4xDP 1.4, VirtualLink (1) | |
| Max Simultaneous Displays | 4x 3840 x 2160 @ 120 Hz, 4x 5120×2880 @ 60 Hz, 2x 7680×4320 @ 60 Hz |
|
| VR Ready | Yes | |
| Graphics APIs | DirectX 12.07, Shader Model 5.17, OpenGL 4.68, Vulkan 1.18 | |
| Compute APIs | CUDA, DirectCompute, OpenCL™ | |
NVIDIA M10 GPU
Great User Experience. High User Density
NVIDIA M10 GPU has Maxwell GPU architecture and works with NVIDIA GRID™ software to provide the industry’s highest user density for virtualized desktops and applications. It supports 64 desktops per board and 128 desktops per server, giving your business the power to deliver great experiences to all of your employees at an affordable cost.
SPECIFICATIONS
| Virtualization Use Case | Density-Optimized Graphics Virtualization | |
| GPU Architecture | NVIDIA Maxwell™ | |
| GPUs per Board | 64 (16 per GPU) | |
| NVIDIA CUDA® Cores | 2560 NVIDIA CUDA Cores (640 per GPU) | |
| GPU Memory | 32 GB of GDDR5 Memory (8 per GPU) | |
| H.264 1080p30 Streams | 28 | |
| Max Power Consumption | 225 W | |
| Thermal Solution | Passive | |
| Form Factor | PCIe 3.0 Dual Slot | |