Try GPU computing power with us!

Rocket
  • 2/4/8 x GeForce GTX 1080
  • Simple registration procedure
  • Dedicated server (very secure!)

VPS with GPU

On the LeaderGPU you can rent high-performance VPS configurations with top hardware (GPU, CPU, memory, etc.) for various calculations. We offer a high-end GPU: FirePro, Titan X, Tesla P100, GeForce GTX 1080.

We provide a high guaranteed bandwidth and a wide range of operating systems. Most operating systems are installed free of charge (exception: Windows Server). The rental price includes 1 IPv4 / IPv6 address.

When renting a VPS with the GPU you have direct access to the GPUs and can use them for any tasks, such as deep learning, rendering, machine learning etc.


TensorFlow benchmark

Server

Example of TensorFlow CIFAR-10 classification on our server with 8 GTX 1080:

  • step 4460, loss = 0.77 (>>>7696.3<<< examples/sec; 0.017 sec/batch)
  • step 4470, loss = 0.91 (>>>7310.9<<< examples/sec; 0.018 sec/batch)
  • step 4480, loss = 1.01 (>>>6577.1<<< examples/sec; 0.019 sec/batch)

Order VPS with the GPU with per-minute, weekly, and monthly payment on the LeaderGPU.

Choose the best solution!

Customer may not use the Services to engage in mining cryptocurrency

  • 2 x 2560 CUDA cores
  • 2 x NVIDIA GeForce GTX1080 TI
  • 22 GB GDDR5x (2 x 11 GB)
  • 2 x Intel Xeon E5-2609v4   1.7 GHz
  • 32 GB RAM
  • 480 GB SSD
  • Ports: 40 Gbps
  • No setup fee
  • 437,34 € / month
  • 109,33 € / week
  • 0,02 € / minute
Order now
  • 4 x 2560 CUDA cores
  • 4 x NVIDIA GeForce GTX1080TI
  • 44 GB GDDR5x TI (4 x 11 GB)
  • 2 x Intel Xeon E5-2609v4   1.7 GHz
  • 64 GB RAM
  • 480 GB SSD
  • Ports: 40 Gbps
  • No setup fee
  • 1 018,38 € / month*
  • 288 € / week*
  • 0,04 € / minute*
Order now

* 4 x GTX 1080 Ti. Special price till end of October 2017!
Standard prices: 1 270 € / month. 317.50 € / week. 0,07 € / minute
Discounted price: 1 018.38 € / month. 288 €/ week. 0,04 € / minute

All Prices exclude VAT.

  • 2 x 2560 CUDA cores
  • 2 x NVIDIA GeForce GTX1080
  • 16 GB GDDR5x (2 x 8 GB)
  • 2 x Intel Xeon E5-2609v4   1.7 GHz
  • 32 GB RAM
  • 480 GB SSD
  • Ports: 40 Gbps
  • No setup fee
  • 364,45 € / month
  • 91,11 € / week
  • 0,02 € / minute
Order now
  • 4 x 2560 CUDA cores
  • 4 x NVIDIA GeForce GTX1080
  • 32 GB GDDR5x (4 x 8 GB)
  • 2 x Intel Xeon E5-2609v4   1.7 GHz
  • 64 GB RAM
  • 480 GB SSD
  • Ports: 40 Gbps
  • No setup fee
  • 848,65 € / month*
  • 240 € / week*
  • 0,03 € / minute*
Order now
  • 8 x 2560 CUDA cores
  • 8 x NVIDIA GeForce GTX1080
  • 64 GB GDDR5x (8 x 8 GB)
  • 2 x Intel Xeon E5-2630v4   2.2 GHz
  • 128 GB RAM
  • 960 GB SSD
  • Ports: 40 Gbps
  • No setup fee
  • 2 017 € / month
  • 504,25 € / week
  • 0,09 € / minute
Order now

* 4 x GeForce GTX 1080. Special price till end of October 2017!
Standard prices: 1 058,33 € / month. 264,58 € / week. 0,06 € / minute
Discounted price: 848.65 € / month. 240 €/ week. 0,03 € / minute

All Prices exclude VAT.

  • 2 x 3584 CUDA cores
  • 2 x NVIDIA PCI Tesla P100
  • 32 GB CoWoS HBM2 (2 x 16 GB)
  • 2 x Intel Xeon E5-2630v4   2.2 GHz
  • 32 GB RAM
  • 480 GB SSD
  • Ports: 40 Gbps
  • No setup fee
  • 1 768,7 € / month
  • 439,68 € / week
  • 0,08 € / minute
Order now

All Prices exclude VAT.

Dedicated Server Features

Guaranteed bandwidth

Internal network (LeaderGPU to LeaderGPU):

  • Included internet traffic: UNLIMITED
  • Burst: 10 Gbps

Internet traffic (LeaderGPU to Internet):

  • Burst internet: 1 Gbps
  • Included internet traffic (monthly based payments): 10 Tb/month
  • Included internet traffic (weekly based payments): 2.5 Tb/week
  • Included internet traffic (minute/hourly based payments): 0 Gb
  • Additional 1Gb (not included): 0,09 €/Gb
Customer service
  • Email: info@leadergpu.com
  • Phone: +31 20 7640722
  • Billing: Monday — Friday: 9 — 6pm
  • Technical Support: Monday — Friday: 9 — 6pm
  • Incident reporting: 24/7
Operating systems
  • Ubuntu Server 16.04 LTS: limited support
  • CentOS 7: free. Access via SSH / WebSSH. Installed drivers, CUDA, cuDNN, python / TensorFlow, etc.
  • Windows Server Standart 2016: free
IPs included
  • 1 IPv4/IPv6 address included
  • Assign additional public IPv4 addresses to your servers — 1€/month per IPv4 address

If you need individual configuration, please make a request and we will contact with you

Send a request

CUDA technology

CUDA — parallel computing architecture designed by NVIDIA. The implementation of this architecture allowed to significantly increasing performance of GPU computing.

Advantages of CUDA:

  • Using the common programming language C
  • Applying of technologies such as Parallel Data Cache and Thread Execution Manager
  • Default libraries for FFT and BLAS
  • Unification of hardware and software parts for GPU computing
  • Multi-GPU Support
  • Hardware Debugging
  • Excellent scalability to different projects
  • Special CUDA driver

OpenCL technology

OpenCL (Open Computing Language) — an open standard, designed for universal parallel programming for different types of processors. OpenCL allows programmers a convenient, flexible and efficient access to heterogeneous computing platforms.

Advantages of OpenCL:

  • Support different techniques of parallelism
  • Uses of a subset of the C99 standard
  • Supports a wide range of software, including the high-performance solutions
  • Independent of the hardware platform and add-on applications
  • Support IEEE 754 standard

We offer GPU AMD FirePro ™ S9300 and FPGA Bittware A10PL4, that support OpenCL technology.

GPU NVIDIA® GeForce GTX 1080

Gtx 1080

Gaming card GTX 1080 is based on Pascal architecture. Designed to address a wide range of computing tasks, thereby applying the solution in 1080 GTX is universal. This concept is perfect for tasks that do not require double precision.


Areas of application servers with the 1080 GTX:


  • Rendering
  • Base machine learning
  • Video Transcoding

  • GPU: GP104 Pascal architecture
  • Memory: 8 GB GDDR5x
  • NVIDIA CUDA cores: 2560
  • Memory Bandwidth: 320 GB/s

GPU NVIDIA NVLink™ Tesla® P100

Tesla p100

NVIDIA NVLink™ Tesla P100 - the most advanced graphics accelerators ever created. They are based on a new architecture GPU NVIDIA Pascal and is the world's fastest computer servers with a capacity exceeding hundreds of classic server-based CPU. If you need high performance and accuracy of calculations - Tesla P 100 is the best choice.


Areas of application servers with the Tesla P 100:


  • 3D modeling
  • Deep learning
  • High-performance computing

  • GPU: P100 Pascal architecture
  • Memory: 16 GB HBM2
  • NVIDIA CUDA cores: 3584
  • Memory Bandwidth: 720 GB/s