1x NVIDIA T4 16GB VRAM 4x vCPUs 6GB RAM
$ 0.35 /hour
1x GPU NVIDIA T4 16GB
4x vCPUs
6GB RAM
16GB VRAM
15GB NVMe workspace volume
Order Now
1x NVIDIA T4 16GB VRAM 8x vCPUs 12GB RAM
$ 0.70 /hour
1x GPU NVIDIA T4 16GB
8x vCPUs
12GB RAM
16GB VRAM
30GB NVMe workspace volume
Order Now
1x NVIDIA T4 16GB VRAM 16x vCPUs 32GB RAM
$ 1.40 /hour
1x GPU NVIDIA T4 16GB
16x vCPUs
32GB RAM
16GB VRAM
60GB NVMe workspace volume
Order Now
T4 servers have:
Up to 8 NVIDIA® T4 GPUs of 16GB each, designed with 2,560 CUDA cores and 320 Tensor cores, this solution delivers versatile performance and power efficiency ideal for AI, inference, and compute graphics workloads.
The architecture Turing™ enables the T4 to achieve an exceptional balance between power and energy consumption, supporting a wide range of applications from machine learning to graphics virtualization.
The Servers optimized for T4 use state-of-the-art processors, ensuring stable and scalable performance to meet the demands of large operations.
When configuring VMs with multiple T4, it is possible to allocate resources flexibly, providing a perfect combination of cost-benefit and high performance.
T4 GPU Applications
AI inference, content recommendation, natural language processing, and video streaming
Streaming
Recommendation systems for e-commerce and streaming.
Server Optimization
Content server optimization for high-quality video streaming.
Chatbots
Implementation of chatbots and virtual assistants.
Nvidia T4 specs
|
|
GPU Architecture |
NVIDIA Turing |
NVIDIA Turing Tensor Cores |
320 |
NVIDIA CUDA® Cores |
2,560 |
Single-Precision |
8.1 TFLOPS |
Mixed-Precision (FP16/FP32) |
65 TFLOPS |
INT8 |
130 TOPS |
INT4 |
260 TOPS |
GPU Memory |
16 GB GDDR6 |
Memory Bandwidth |
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 |
1x NVIDIA V100 16GB VRAM 4x vCPUs 6GB RAM
$ 0.65 /hour
1x GPU NVIDIA V100 16GB
4x vCPUs
6GB RAM
16GB VRAM
15GB NVMe workspace volume
Order Now
1x NVIDIA V100 16GB VRAM 8x vCPUs 12GB RAM
$ 1.30 /hour
1x GPU NVIDIA V100 16GB
8x vCPUs
12GB RAM
16GB VRAM
30GB NVMe workspace volume
Order Now
1x NVIDIA V100 16GB VRAM 16x vCPUs 32GB RAM
$ 2.60 /hour
1x GPU NVIDIA V100 16GB
16x vCPUs
32GB RAM
16GB VRAM
60GB NVMe workspace volume
Order Now
V100 servers have:
Up to 8 NVIDIA® Tesla V100 GPUs, each designed with 5120 CUDA cores and 640 Tensor cores for exceptional performance.
Intel Xeon Scalable 4214R second-generation processors [2020], offering up to 48 threads and a 3.5 GHz boost frequency.
Includes NVLink technology, providing high-bandwidth P2P communication. Servers optimized for V100 use state-of-the-art processors, ensuring stable and scalable performance to meet the demands of large operations.
When configuring VMs with multiple V100, it is possible to allocate resources flexibly, providing a perfect combination of cost-benefit and high performance.
V100 GPU Applications
AI training and inference, large-scale data analytics, and deep learning
Data Analytics
Advanced analytics on large datasets in financial and research sectors.
Medical Diagnostics
Developing AI for medical diagnosis with images.
ML and DL Training
Training machine learning and deep learning models.
Nvidia V100 specs
|
|
GPU Architecture |
NVIDIA Volta |
NVIDIA Turing Tensor Cores |
640 |
NVIDIA CUDA® Cores |
5,120 |
Double-Precision Performance |
7 TFLOPS |
Single-Precision Performance |
14 TFLOPS |
Tensor Performance |
112 TFLOPS |
GPU Memory |
32 GB / 16 GB HBM2 |
Memory Bandwidth |
900 GB/sec |
ECC |
Yes |
Interconnect Bandwidth |
32 GB/sec |
System Interface |
PCIe Gen3 |
Form Factor |
PCIe Full Height/Length |
Max Power Consumption |
250 W |
Thermal Solution |
Passive |
Compute APIs |
CUDA, DirectCompute, OpenCL™, OpenACC® |
1x NVIDIA A100 40GB
$ 5.14 /hour
1x GPU NVIDIA A100 40GB
4x vCPUs
6GB RAM
40GB VRAM
15GB NVMe workspace volume
1x NVIDIA A100 80GB
$ 8.52 /hour
1x GPU NVIDIA A100 80GB
4x vCPUs
6GB RAM
80GB VRAM
15GB NVMe workspace volume
A100 servers have:
Up to 8 80 GB NVIDIA® A100 GPUs, each with 6,912 CUDA cores and 432 Tensor cores.
We exclusively use the SXM4 module, designed for NVLINK, which provides memory bandwidth greater than 2 TB/ s and a P2P bandwidth of up to 600 GB/s.
Second AMD EPYC Rome Processor generation, with support for up to 192 threads and a maximum frequency of 3.3 GHz in boost mode.
When configuring VMs with multiple T4, it is possible to allocate resources flexibly, providing a perfect combination of cost- benefit and high performance.
A100 GPU Applications
Next-generation AI training, HPC, deep learning, and mixed training and inference workloads
Predictive Analysis
Developing robust AI models for forecasting, simulation, and predictive analysis.
Scientific Simulations
Scientific simulations for drug development and research.
Bioinformatics
Applications in bioinformatics, such as genomic sequencing.
Nvidia A100 specs
|
|
FP64 |
9.7 TFLOPS |
FP64 Tensor Core |
19.5 TFLOPS |
FP32 |
19.5 TFLOPS |
Tensor Float 32 (TF32) |
156 TFLOPS, 312 TFLOPS |
BFLOAT16 Tensor Core |
312 TFLOPS, 624 TFLOPS |
FP16 Tensor Core |
312 TFLOPS, 624 TFLOPS |
INT8 Tensor Core |
624 TOPS, 1248 TOPS |
GPU Memory |
80GB HBM2e |
GPU Memory Bandwidth |
1,935GB/s |
Max Thermal Design Power (TDP) |
300W |
Multi-Instance GPU |
Up to 7 MIGs @ 10GB |
Form Factor |
PCIe |
Interconnect |
NVIDIA® NVLink® Bridge for 2 GPUs: 600GB/s |
PCIe Gen4 |
64GB/s |
Server Options |
Partner and NVIDIA-Certified Systems™ with 1-8 GPUs |
1x NVIDIA H100 80GB
$ 17.77 /hour
1x GPU NVIDIA H100 80GB
4x vCPUs
6GB RAM
80GB VRAM
15GB NVMe workspace volume
H100 servers have:
Up to 8 80GB NVIDIA® H100 GPUs, each equipped with 16,896 CUDA cores and 528 Tensor cores.
These GPUs represent the top of NVIDIA's current lineup, delivering unmatched performance for artificial intelligence applications.
The SXM5 NVLINK module, which achieves a memory bandwidth of 2.6 Gbps and enables peer-to-peer communication (P2P) with speeds of up to 900 GB/s.
Fourth generation AMD Genoa processors with up to 384 threads, operating at a maximum frequency of 3.7 GHz.
The 8H100 configuration uses the same number of vCPUs than the 4H100, but stands out in terms of CPU performance by taking advantage of only physical cores, eliminating the need for hyper-threads.
H100 GPU Applications
Training giant AI models, massive simulations, generative AI, and HPC
RT Analysis
AI for real-time analysis, such as in autonomous driving and security systems.
Language Simulation
AI for content creation and large-scale natural language simulation.
Advanced Research
Advanced scientific research and complex simulations, such as in particle physics.
Nvidia H100 specs
|
|
FP64 |
26 TFLOPS |
FP64 Tensor Core |
51 TFLOPS |
FP32 |
51 TFLOPS |
TF32 Tensor Core |
756 TFLOPS |
BFLOAT16 Tensor Core |
1,513 TFLOPS |
FP16 Tensor Core |
1,513 TFLOPS |
FP8 Tensor Core |
3,026 TFLOPS |
INT8 Tensor Core |
3,026 TOPS |
GPU Memory |
80GB |
GPU Memory Bandwidth |
2TB/s |
Decoders |
7 NVDEC, 7 JPEG |
Max Thermal Design Power (TDP) |
300-350W |
Multi-Instance GPUs |
Up to 7 MIGs @ 10GB each |
Form Factor |
PCIe, dual-slot, air-cooled |
1x NVIDIA L4 24GB
$ 1.80 /hour
1x GPU NVIDIA L4 24GB
4x vCPUs
6GB RAM
24GB VRAM
15GB NVMe workspace volume
L4 servers have:
Up to 8 NVIDIA® L4 GPUs at 48 GB each, offering 18,176 CUDA cores and 568 Tensor cores per unit.
Memory bandwidth reaches 864 GB/s with 1466 Tflops Tensor performance capacity, 212 Tflops RT Core and 91.6 Simple Accuracy Tflops.
L4-optimized servers utilize state-of-the-art processors, ensuring stable and scalable performance to meet the demands of large operations.
When configuring multiple L4 VMs, it is It is possible to allocate resources flexibly, providing a perfect combination of cost-benefit and high performance.
L4 GPU Applications
Inference and video acceleration, AI applications optimized for low power consumption
Streaming Application
Streaming and live video transcoding applications.
Systems and Devices
AI applied to low-power devices, such as IoT systems and embedded devices.
Video Processing
Video and image processing for smart security cameras.
Nvidia L4 specs
|
|
FP32 |
30.3 TFLOPs |
TF32 Tensor Core |
120 TFLOPs |
FP16 Tensor Core |
242 TFLOPs |
BFLOAT16 Tensor Core |
242 TFLOPs |
FP8 Tensor Core |
485 TFLOPs |
INT8 Tensor Core |
485 TOPs |
GPU Memory |
24GB |
GPU Memory Bandwidth |
300 GB/s |
NVENC, NVDEC, JPEG Decoders |
2, 4, 4 |
Max Thermal Design Power (TDP) |
72W |
Form Factor |
1-slot low-profile, PCIe |
Interconnect |
PCIe Gen4 x16 64GB/s |
Server Options |
Partner and NVIDIA-Certified Systems with 1-8 GPUs |
RTX 3090 servers have:
Up to 8 NVIDIA® RTX 3090 GPUs, each equipped with 10,496 CUDA cores and 328 Tensor cores, designed to deliver extreme performance in intensive applications.
With high-performance Intel Xeon Scalable processors, up to 48 threads can be achieved and boost frequencies optimized for demanding workloads.
NVLink technology ensures direct communication between GPUs with bandwidth high, ideal for parallel workloads and processing large volumes of data.
Servers optimized for the RTX 3090 enable flexible configurations, combining scalability and efficiency to meet the most varied needs, from 3D modeling to training State-of-the-art AI.
RTX 3090 GPU Applications
AI training and inference on a smaller scale, video editing, animation editing, and 3D rendering
AI for Startups
Training AI models in startups and development environments.
Games and Simulations
Virtual reality applications in game development and simulations.
Rendering
Content production, 3D rendering, video editing, and graphic design.
Nvidia RTX 3090 specs
|
|
GPU Architecture |
Ampere |
Process Size |
8 nm |
Transistors |
28.3 billion |
Bus Interface |
PCIe 4.0 x16 |
Base Clock |
1.40 GHz |
Boost Clock |
1.70 GHz |
Memory Clock |
19.5 Gbps |
Memory Size |
24 GB GDDR6X |
Memory Type |
GDDR6X |
Memory Bus |
384-bit |
Bandwidth |
936.2 GB/s |
Tensor Cores |
328 |
FP32 |
39.7 TFLOPS |
FP64 |
1.25 TFLOPS |
CUDA Cores |
10,496 |
RTX 4090 servers have:
Up to 8 NVIDIA® RTX 4090 GPUs, each equipped with 16,384 CUDA cores and 512 Tensor cores, designed to deliver cutting-edge performance in high-demand applications such as artificial intelligence, simulations, and advanced rendering.
With support for Intel Xeon Scalable processors, the servers offer up to 48 threads and optimized frequencies to ensure maximum efficiency.
NVLink technology enables communication direct between GPUs, offering high bandwidth for parallel operations and large-scale data transfers.
Servers optimized for the RTX 4090 provide flexible configurations, combining high performance and energy efficiency to meet the demands of professionals who seek robust and scalable solutions.
RTX 4090 GPU Applications
AI and Machine Learning applications on a smaller scale, game development, simulations, and virtual reality
AI Prototyping
Testing and prototyping AI models in early research stages.
Interactive Simulations
Development of interactive simulations in training and gaming environments.
Video Editing
Animation development, video editing, and special effects.
Nvidia RTX 4090 specs
|
|
Model Name |
GeForce RTX™ 4090 GAMING X TRIO 24G |
Graphics Processing Unit |
NVIDIA® GeForce RTX™ 4090 |
Interface |
PCI Express® Gen 4 |
Core Clocks |
2610 MHz, Boost: 2595 MHz |
CUDA® CORES |
16,384 Units |
Memory Speed |
21 Gbps |
Memory |
24GB GDDR6X |
Memory Bus |
384-bit |
Output |
DisplayPort x 3 (v1.4a), HDMI™ x 1 |
HDCP Support |
Yes |
Power Consumption |
450 W |
Power Connectors |
16-pin x 1 |
Recommended PSU |
850 W |
Card Dimensions (mm) |
337 x 140 x 77 mm |
Weight (Card / Package) |
2170 g / 3093 g |
DirectX Version Support |
12 Ultimate |
OpenGL Version Support |
4.6 |
Maximum Displays |
4 |
G-SYNC® Technology |
Yes |
Digital Maximum Resolution |
7680x4320 |
RTX 4000 servers have:
Up to 8 NVIDIA® RTX 4000 GPUs, each equipped with 7,680 CUDA cores and 240 Tensor cores, designed to deliver superior performance in advanced graphics and data processing.
With support for Intel Xeon Scalable processors, these systems deliver up to 48 threads and frequencies optimized to meet the demands of high-performance applications.
NVLink technology enables efficient communication between GPUs, offering high bandwidth to maximize productivity in parallel workloads.
Servers configured with RTX 4000 GPUs are ideal for applications such as graphic design, rendering, simulations and artificial intelligence, combining flexibility and performance to demanding corporate environments.
RTX 4000 GPU Applications
High-performance graphics computing, visualization, design, and 3D modeling
Realistic Visualization
Creating realistic visualizations for architecture and construction.
Detailed Applications
Scientific and technical visualization applications with detailed and accurate graphics.
Design Environments
Product design, engineering, and CAD simulation environments.
Nvidia RTX 4000 specs
|
|
GPU Memory |
20GB GDDR6 |
Memory Interface |
160-bit |
Memory Bandwidth |
280 GB/s |
Error Correcting Code |
Yes |
Architecture |
NVIDIA Ada Lovelace |
CUDA Cores |
6,144 |
Tensor Cores (Fourth-generation) |
192 |
RT Cores (Third-generation) |
48 |
Single-Precision Performance |
19.2 TFLOPS |
RT Core Performance |
44.3 TFLOPS |
Tensor Performance |
306.8 TFLOPS |
System Interface |
PCIe 4.0 x 16 |
Power Consumption |
Total board power: 70 W |
Thermal Solution |
Active |
Form Factor |
2.7” H x 6.6” L, dual slot |
Encode/Decode Engines |
2x encode, 2x decode (+AV1 encode and decode) |
VR Ready |
Yes |
vGPU Software Support |
No |
Graphics APIs |
DirectX 12, Shader Model 6.6, OpenGL 4.65, Vulkan 1.35 |
Compute APIs |
CUDA 11.6, OpenCL 3.0, DirectCompute |
NVIDIA NVLink® |
No |