immers.cloud platform
All GPU servers are based on Intel® Xeon® Scalable 2nd, 3rd and 5th generation processors and contain up to 112 virtual cores and up to 8192 GB of DDR5 ECC Reg 5200 MHz RAM.
Each processor is equipped with two Intel® AVX-512 units and supports Intel® AVX-512 Deep Learning Boost functions, which accelerate multiplication and addition operations with reduced accuracy, enhancing performance in deep learning algorithms.
Local storage is organized using Intel® and Samsung solid-state drives designed specifically for data centers, with a capacity of up to 7,68 TB.
100% Performance
Each physical core or GPU adapter is dedicated to a single client.
This means:
- 100% vCPU time is available;
- Physical pass-through of GPUs inside virtual servers;
- Reduced storage and network load on hypervisors, delivering more storage and network performance to clients.
Up to 75,000 IOPS1 for RANDOM READ and up to 20,000 IOPS for RANDOM WRITE on Virtual Servers with local SSDs
Up to 70 000 IOPS1 for RANDOM READ and up to 60 000 IOPS for RANDOM WRITE on Virtual Servers with block storage volumes
You can be confident that Virtual Servers do not share vCPU or GPU resources with one another.
GPU NVIDIA H200 141 GB
First GPU surpass 100GB of VRAM on a single chip.
GPU NVIDIA® H100 NVL 94 GB
Update of the popular H100 GPU with increased memory capacity and bandwidth.
GPU NVIDIA® H100 80 GB
NVIDIA H100 GPU provides unsurpassed acceleration for AI tasks, data analysis and for solving the most complex computing tasks.
GPU RTX 5090
Blackwell architecture, based on a updated 5 nm process technology, provides a maximum performance in FP4 and DLSS4 and now optimized for neural shaders. Each accelerator has 21760 CUDA cores, and 32 GB of GDDR7 memory.
GPU RTX 4090
Ada Lovelace architecture, based on a new 5 nm process technology, provides a huge leap in performance, efficiency and graphics. Each accelerator has 16384 CUDA cores, and 24 GB of GDDR6X memory.
GPU RTX™ 3090
RTX 3090 graphics cards are based on the powerful Ampere architecture and a improved RTX hardware ray tracing platform. Each accelerator has 328 tensor cores, 10496 CUDA cores, and 24 GB of memory.
GPU RTX™ 3080
RTX 3080 graphics cards are based on the powerful Ampere architecture and a improved RTX hardware ray tracing platform. Each accelerator has 272 tensor cores, 8704 CUDA cores, and 10 GB of memory.
GPU Tesla® A100 80 GB
Tesla A100 GPU provides unsurpassed acceleration for AI tasks, data analysis and for solving the most complex computing tasks. The A100 is the most productive integrated platform for AI and HPC, allowing you to get real-time results and deploy scalable solutions.
GPU RTX® A5000
The RTX A5000 graphics accelerator has the perfect balance of power, performance and reliability to solve complex tasks. This GPU is built on the basis of the latest Ampere architecture and has 24 GB of video memory — everything so that designers, engineers and artists can implement the projects they dreamed of today.
Thanks to the new CUDA cores, which provide up to 2.5 times FP32 performance compared to the previous generation, work with graphics is accelerated.
Higher rendering accuracy is provided by hardware acceleration of motion blur and higher ray tracing performance.
In flavors with an even number of GPUs, graphics adapters are combined using NVLink, which allows you to increase the amount of memory and improve performance for performing complex visual calculations.
GPU Tesla® A10
Tesla A10 graphics accelerators, featuring tensor cores, are built on the Ampere architecture, which enhances performance and efficiency for various computing tasks.
Thanks to CUDA cores, the Tesla A10 accelerators deliver twice the number of single-precision floating-point operations (FP32) compared to previous generations, significantly speeding up work with graphics, video, and modeling complex 3D models in computer-aided design (CAD) software.
The second generation of RT cores enables simultaneous ray tracing, shading, or noise reduction, accelerating tasks such as photorealistic rendering of film materials, architectural project evaluation, and motion rendering for faster, more accurate results.
Support for Tensor Float 32 (TF32) operations in Tesla A10 accelerators boosts training speeds for AI models and data processing by five times compared to previous generations, without requiring changes in the code. Tensor cores also enable AI-based technologies such as DLSS, noise reduction, and photo and video editing functions in select applications.
PCI Express Gen 4 doubles the bandwidth of PCIe Gen 3, accelerating data transfer from processor memory for resource-intensive tasks like AI, data processing, and 3D graphics rendering.
Thanks to ultra-fast GDDR6 memory, scientists, engineers, and data science specialists gain the necessary resources for processing large datasets and conducting advanced modeling.
GPU RTX™ 2080 Ti
RTX 2080 Ti graphics cards are based on the powerful Turing architecture and a completely new RTX hardware ray tracing platform. Each accelerator has 544 2nd gen tensor cores, 4352 CUDA cores, and 11 GB of memory.
GPU Tesla® A2
The Tesla A2 graphics accelerator is optimized for inference tasks and provides up to 1.3 times greater performance for smart cities, industry and retail tasks.
GPU Tesla® T4
Tesla® T4 with tensor and RT cores is the one of most advanced and energy-efficient graphics accelerator for inference, video transcoding, streaming, and remote desktops.
Each accelerator has 320 tensor cores, 2560 CUDA cores, and 16 GB of memory.
T4 graphics accelerators are ideal for operating neural network models in a production environment (inferencing), speech processing, and NLP.
In addition to tensor cores, T4 has RT cores that perform hardware ray tracing (retrace).
GPU Tesla® V100
Tesla® V100 with tensor cores is the first accelerator to break the 100 teraoperations per second (TOPS) performance barrier in deep learning tasks.
Each graphics accelerator has 640 tensor cores, 5120 CUDA cores, and 32 GB of HBM2 memory with a maximum throughput of 900 GB/s.
The total computing performance of the server is 28 TFLOPS with double precision and 130 TFLOPS with mixed-precision and tensor cores.
V100 accelerators are ideal for training deep neural networks.
GPU server hosting: power for your compute-intensive workloads
At immers.cloud, you can host a server with an NVIDIA GPU — from consumer-grade RTX 4090/5090 to professional A100, H100, and H200 accelerators. All configurations come with fully isolated resources: you get 100% dedicated CPU cores and GPU capacity: no overselling, no performance throttling.
Get up to 50% off with a 6-month pre-paid plan.
Register now
What are GPU servers used for?
| Machine Learning & AI |
Training and inference of large language models (LLMs), working with transformers, fine-tuning, and agent-based systems |
| Image Generation & Multimodal Models |
Running diffusion models such as Stable Diffusion, FLUX, LTX, and others with high-resolution output and batch processing |
| Rendering & 3D Graphics |
Hardware-accelerated ray tracing (RT cores), rendering in Blender, Maya, Unreal Engine, and other professional software |
| Video Transcoding & Streaming |
Hardware encoding/decoding via NVENC/NVDEC, real-time live streaming, and adaptive video delivery for various devices |
| Scientific Computing & HPC |
Parallel CUDA computations, simulations, data analysis, bioinformatics, computational fluid dynamics (CFD), and other compute-intensive workloads |
| Virtual Workstations |
Remote access to graphics-intensive applications such as AutoCAD, Adobe Creative Suite, and DaVinci Resolve with low latency and full GPU acceleration |
Ready-made OS images with the required software
Create virtual servers by utilizing our pre-configured OS images with either Windows or Linux, along with specialized pre-installed software.
View all the pre-installed images
in the Marketplace.
Pure OpenStack API
Developers and system administrators can manage the cloud using the full OpenStack API.
Authenticate ninja_user example:
$ curl -g -i -X POST https://api.immers.cloud:5000/v3/auth/tokens \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-H "User-Agent: YOUR-USER-AGENT" \
-d '{"auth": {"identity": {"methods": ["password"], "password": {"user": { "name": "ninja_user", "password": "ninja_password", "domain": {"id": "default"}}}}, "scope": {"project": {"name": "ninja_user", "domain": {"id": "default"}}}}}'
Create ninja_vm example:
$ curl -g -i -X POST https://api.immers.cloud:8774/v2.1/servers \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-H "User-Agent: YOUR-USER-AGENT" \
-H "X-Auth-Token: YOUR-API-TOKEN" \
-d '{"server": {"name": "ninja_vm", "imageRef": "8b85e210-d2c8-490a-a0ba-dc17183c0223", "key_name": "mykey01", "flavorRef": "8f9a148d-b258-42f7-bcc2-32581d86e1f1", "max_count": 1, "min_count": 1, "networks": [{"uuid": "cc5f6f4a-2c44-44a4-af9a-f8534e34d2b7"}]}}'
STOP ninja_vm example:
$ curl -g -i -X POST https://api.immers.cloud:8774/v2.1/servers/{server_id}/action \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-H "User-Agent: YOUR-USER-AGENT" \
-H "X-Auth-Token: YOUR-API-TOKEN" \
-d '{"os-stop" : null}'
START ninja_vm example:
$ curl -g -i -X POST https://api.immers.cloud:8774/v2.1/servers/{server_id}/action \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-H "User-Agent: YOUR-USER-AGENT" \
-H "X-Auth-Token: YOUR-API-TOKEN" \
-d '{"os-start" : null}'
SHELVE ninja_vm example:
$ curl -g -i -X POST https://api.immers.cloud:8774/v2.1/servers/{server_id}/action \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-H "User-Agent: YOUR-USER-AGENT" \
-H "X-Auth-Token: YOUR-API-TOKEN" \
-d '{"shelve" : null}'
DELETE ninja_vm example:
$ curl -g -i -X DELETE https://api.immers.cloud:8774/v2.1/servers/{server_id} \
-H "User-Agent: YOUR-USER-AGENT" \
-H "X-Auth-Token: YOUR-API-TOKEN"
Documentation
Subscribe to our newsletter
Get notifications about new promotions and special offers by email.
