GPU Servers Rental

NVIDIA GeForce GTX 1080 Ti

Nvidia GeForce GTX 1080 Ti is based on the Pascal GP102 graphics processor (the same chip used by Tesla P40)

GeForce GTX 1080 Ti GPUs are good for Deep Learning, animation rendering, video transcoding and real time video processing.

11 GB memory (484 GB/s bandwidth)
3584 Nvidia CUDA cores
11.3 TFLOPS (FP32)

NVIDIA GeForce RTX 2080 Ti

Nvidia GeForce RTX 2080 Ti is based on the TU102 graphics processor (the same processor used by Titan RTX). In addition to CUDA cores, RTX 2080 Ti has 544 tensor cores that accelerates machine learning applications. The card also has 68 ray tracing cores which dramatically improves GPU ray tracing performance and it's supported in VRAY, OctaneRender and Redshift.

GeForce RTX 2080 Ti GPUs are good for Deep Learning, animation rendering, real time ray tracing, video transcoding and VR/AR design.

11 GB memory (616 GB/s bandwidth)
4352 Nvidia CUDA cores
13.5 TFLOPS (FP32)

NVIDIA Tesla V100

Nvidia Tesla V100 is based on Nvidia Volta architecture and offers the performance of up to 100 CPUs in a single GPU. V100 has two versions, PCle and SXM2, with 14 and 15.7 TFLOPS FP32 performance respectively. It comes with either 16GB or 32GB memory, and 32GB one enables AI teams to fit very large neural network models into the GPU memory.

Nvidia Tesla V100 GPUs are good for Machine Learning, Deep Learning, Natural Language Processing, molecular modeling and genomics. Tesla GPUs are the only choice for applications that require double precision such as physics modelling or engineering simulations.

16 or 32 GB memory (900 GB/s bandwidth)
5120 Nvidia CUDA cores
15.7 TFLOPS (FP32)

NVIDIA RTX 3090

Nvidia RTX 3090 enthusiast-class graphics card and is based on Nvidia Ampere architecture (GA102-300-A1 variant).

24 GB memory (936 GB/s bandwidth)
10496 Nvidia CUDA cores
328 Tensor Cores
35.5 TFLOPS (FP32)

NVIDIA RTX A6000

Nvidia RTX A6000 is based on Nvidia Ampere architecture and empowers designers, engineers, scientists, and artists to handle the most graphics and compute-intensive workflows. The RTX A6000 is equipped with the latest generation RT Cores, Tensor Cores, and CUDA cores for unprecedented rendering, AI, graphics, and compute performance.

New Tensor Float 32 (TF32) precision provides up to 5X the training throughput over the previous generation to accelerate AI and data science model training without requiring any code changes. Hardware support for structural sparsity doubles the throughput for inferencing.

NVIDIA Ampere architecture based GPUs support PCI Express Gen 4.0 (PCIe 4.0), which provides 2X the bandwidth of PCIe Gen 3.0. This improves data transfer speeds from CPU memory for data-intensive tasks such as AI and data science. Faster PCIe performance also accelerates GPU direct memory access (DMA) transfers, enabling faster video data transfers from GPUDirect® for video-enabled devices.

Third-generation NVIDIA NVLink® technology enables users to connect two GPUs together to share GPU performance and memory. With up to 112 gigabytes per second (GB/s) of bidirectional bandwidth and combined graphics memory of up to 96 GB, professionals can tackle the largest rendering, AI, virtual reality, and visual computing workloads.

48 GB memory (768 GB/s bandwidth)
10752 Nvidia CUDA cores
336 Tensor Cores
38.7 TFLOPS (FP32)
309.7 TFLOPS Tensor Performance

NVIDIA RTX A4000

Nvidia RTX A4000 is the most powerful single-slot GPU for professionals. Built on the NVIDIA Ampere architecture, the RTX A4000 come with 16 GB of graphics memory with error-correction code (ECC). Our A4000 based servers provides the highest performance/dollar.

16 GB memory (448 GB/s bandwidth)
6144 Nvidia CUDA cores
192 Tensor Cores
19.2 TFLOPS (FP32)
153.4 TFLOPS Tensor Performance