Deploy & Scale Phi AI Models on High-Performance
Run, Fine-Tune & Serve Phi LLMs on Dedicated GPU Infrastructure with Fast, Reliable AI Inference.
Eight-Core Xeon E5-2690 |
120GB + 960GB SSD | 100Mbps-1Gbps |
-
Nvidia Quadro P1000 |
32GB RAM | GPU -
Windows / Linux | Microarchitecture: Pascal |
OS -
640 | GPU Memory: 4GB GDDR5 |
CUDA Cores -
1.894 TFLOPS
FP32 Performance
Eight-Core Xeon E5-2690 |
120GB + 960GB SSD | 100Mbps-1Gbps |
-
Nvidia Quadro T1000 |
64GB RAM | GPU -
Windows / Linux | Microarchitecture: Turing |
OS -
896 | GPU Memory: 8GB GDDR6 |
CUDA Cores -
2.5 TFLOPS
FP32 Performance
Eight-Core Xeon E5-2667v3 |
120GB + 960GB SSD | 100Mbps-1Gbps |
-
Nvidia GeForce GTX 1650 |
64GB RAM | GPU -
Windows / Linux | Microarchitecture: Turing |
OS -
896 | GPU Memory: 4GB GDDR5 |
CUDA Cores -
3.0 TFLOPS
FP32 Performance
Dual 8-Core Xeon E5-2660 | 120GB + 960GB SSD |
-
Nvidia GeForce GTX 1660 |
64GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Turing | CUDA Cores: 1408 |
Microarchitecture -
6GB GDDR6 | FP32 Performance: 5.0 TFLOPS
GPU Memory
Dual 12-Core E5-2690v3 | 240GB SSD + 2TB SSD |
-
Nvidia V100 |
128GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Volta | CUDA Cores: 5,120 |
Microarchitecture -
640 | GPU Memory: 16GB HBM2 |
Tensor Cores -
14 TFLOPS
FP32 Performance
Dual 8-Core E5-2660 |
120GB + 960GB SSD | 100Mbps-1Gbps |
-
Nvidia GeForce RTX 2060 |
128GB RAM | GPU -
Windows / Linux | Microarchitecture: Ampere |
OS -
1,920 | Tensor Cores: 240 |
CUDA Cores -
6GB GDDR6 | FP32 Performance: 6.5 TFLOPS
GPU Memory
Dual 20-Core Gold 6148 | 120GB + 960GB SSD |
-
Nvidia GeForce RTX 2060 |
128GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Ampere | CUDA Cores: 1,920 |
Microarchitecture -
240 | GPU Memory: 6GB GDDR6 |
Tensor Cores -
6.5 TFLOPS
FP32 Performance
Dual 12-Core E5-2697v2 | 240GB SSD + 2TB SSD |
-
GeForce RTX 3060 Ti |
128GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Ampere |
Microarchitecture -
4,864 | Tensor Cores: 152 |
CUDA Cores -
8GB GDDR6 | FP32 Performance: 16.2 TFLOPS
GPU Memory
30GB RAM | 24 CPU Cores | 320GB SSD |
300Mbps Unmetered Bandwidth |
Once per 2 Weeks Backup |
-
Windows / Linux |
OS -
Quadro RTX A4000 |
Dedicated GPU -
6,144 | Tensor Cores: 192 |
CUDA Cores -
16GB GDDR6 |
GPU Memory -
19.2 TFLOPS
FP32 Performance
Dual 12-Core E5-2697v2 | 240GB SSD + 2TB SSD |
-
Nvidia Quadro RTX A4000 |
128GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Ampere |
Microarchitecture -
6,144 | Tensor Cores: 192 |
CUDA Cores -
16GB GDDR6 | FP32 Performance: 19.2 TFLOPS
GPU Memory
Dual 12-Core E5-2697v2 |
240GB SSD + 2TB SSD | 100Mbps-1Gbps |
-
Nvidia Quadro RTX A5000 |
128GB RAM | GPU -
Windows / Linux | Microarchitecture: Ampere |
OS -
8,192 | Tensor Cores: 256 |
CUDA Cores -
24GB GDDR6 | FP32 Performance: 27.8 TFLOPS
GPU Memory
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
Nvidia A40 |
256GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Ampere | CDA Cores: 10,752 |
Microarchitecture -
336 | GPU Memory: 48GB GDDR6 |
Tensor Cores -
37.48 TFLOPS
FP32 Performance
28GB RAM | 16 CPU Cores | 240GB SSD |
200Mbps Unmetered Bandwidth |
-
Windows / Linux |
Once per 4 Weeks Backup | OS -
Nvidia GeForce RTX 5060 |
GPU -
4,608 | Tensor Cores: 144 |
CUDA Cores -
8GB GDDR7 | FP32 Performance: 23.22 TFLOPS
GPU Memory
28GB RAM | 16 CPU Cores | 240GB SSD |
300Mbps Unmetered Bandwidth |
Once per 2 Weeks Backup |
-
Windows / Linux |
OS -
Nvidia RTX Pro 2000 |
Dedicated GPU -
4,352 | Tensor Cores: 5th Gen |
CUDA Cores -
16GB GDDR7 | FP32 Performance: 17 TFLOPS
GPU Memory
24-Core Platinum 8160 | 120GB SSD + 960GB SSD |
-
Nvidia GeForce RTX 5060 |
64GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Blackwell 2.0 |
Microarchitecture -
4,608 | Tensor Cores: 144 |
CUDA Cores -
8GB GDDR7 | FP32 Performance: 23.22 TFLOPS
GPU Memory
60GB RAM | 24 CPU Cores | 320GB SSD |
500Mbps Unmetered Bandwidth |
-
Windows / Linux |
Once per 2 Weeks Backup | OS -
Nvidia RTX Pro 4000 |
Dedicated GPU -
8,960 | Tensor Cores: 280 |
CUDA Cores -
24GB GDDR7 | FP32 Performance: 34 TFLOPS
GPU Memory
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
GeForce RTX 5090 |
256GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Blackwell 2.0 |
Microarchitecture -
21,760 | Tensor Cores: 680 |
CUDA Cores -
32GB GDDR7 | FP32 Performance: 109.7 TFLOPS
GPU Memory
60GB RAM | 24 CPU Cores | 320GB SSD |
500Mbps Unmetered Bandwidth |
Once per 2 Weeks Backup |
-
Windows / Linux |
OS -
Nvidia RTX Pro 5000 |
Dedicated GPU -
14,080 | Tensor Cores: 440 |
CUDA Cores -
48GB GDDR7 | FP32 Performance: 66.94 TFLOPS
GPU Memory
90GB RAM | 32 CPU Cores | 400GB SSD |
1000Mbps Unmetered Bandwidth |
Once per 2 Weeks Backup |
-
Windows / Linux |
OS -
Nvidia RTX Pro 6000 |
Dedicated GPU -
24,064 | Tensor Cores: 852 |
CUDA Cores -
96GB GDDR7 | FP32 Performance: 126 TFLOPS
GPU Memory
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
Nvidia A100 |
256GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Ampere | CUDA Cores: 6,912 |
Microarchitecture -
432 | GPU Memory: 80GB HBM2e |
Tensor Cores -
19.5 TFLOPS
FP32 Performance
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
Nvidia H100 |
256GB RAM | GPU -
Windows / Linux |
100Mbps-1Gbps | OS -
Hopper | CUDA Cores: 14,592 |
Microarchitecture -
456 | GPU Memory: 80GB HBM2e |
Tensor Cores -
183 TFLOPS
FP32 Performance
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
2 × GeForce RTX 4090 |
256GB RAM | GPU -
Windows / Linux |
1Gbps | OS -
Ada Lovelace |
Microarchitecture -
16,384 | Tensor Cores: 512 |
CUDA Cores -
24GB GDDR6X | FP32 Performance: 82.6 TFLOPS
GPU Memory
Dual E5-2699v4 | 240GB SSD + 2TB NVMe + 8TB SATA |
-
2 × GeForce RTX 5090 |
256GB RAM | GPU -
Windows / Linux |
1Gbps | OS -
Blackwell 2.0 |
Microarchitecture -
21,760 | Tensor Cores: 680 |
CUDA Cores -
32GB GDDR7 | FP32 Performance: 109.7 TFLOPS
GPU Memory
Dual 18-Core E5-2697v4
240GB SSD + 2TB NVMe + 8TB SATA | 1Gbps |
-
3 × Nvidia V100 |
256GB RAM | GPU -
Windows / Linux | Microarchitecture: Volta
OS -
5,120 | Tensor Cores: 640 |
CUDA Cores -
16GB HBM2 | FP32 Performance: 14 TFLOPS
GPU Memory
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
3 × Quadro RTX A5000 |
256GB RAM | GPU -
Windows / Linux |
1Gbps | OS -
Ampere |
Microarchitecture -
8,192 | Tensor Cores: 256 |
CUDA Cores -
24GB GDDR6 | FP32 Performance: 27.8 TFLOPS
GPU Memory
Dual 18-Core E5-2697v4 |
240GB SSD + 2TB NVMe + 8TB SATA |
-
3 × Quadro RTX A6000 |
256GB RAM | GPU -
Windows / Linux |
1Gbps | OS -
Ampere | CUDA Cores: 10,752
Microarchitecture -
336 | GPU Memory: 48GB GDDR6 |
| Tensor Cores -
38.71 TFLOPS
FP32 Performance
Dual 22-Core E5-2699v4 |
240GB SSD + 4TB NVMe + 16TB SATA |
-
4 × Nvidia A100 |
512GB RAM | GPU -
Windows / Linux |
1Gbps | OS -
Ampere |
Microarchitecture -
6,912 | Tensor Cores: 432 |
CUDA Cores -
40GB HBM2 |
GPU Memory -
19.5 TFLOPS
FP32 Performance
Dual 22-Core E5-2699v4 |
240GB SSD + 4TB NVMe + 16TB SATA |
-
4 × Quadro RTX A6000 |
512GB RAM | GPU -
Windows / Linux |
1Gbps | OS -
Ampere |
Microarchitecture -
10,752 | Tensor Cores: 336 |
CUDA Cores -
GPU Memory
