Trooper.AI delivers NVIDIA RTX 4090 24 GB & NVIDIA RTX 4090 Pro 48 GB GPU servers with full power and zero overhead. Whether you’re training large models, running complex AI inference, or fine-tuning LLMs – our Ada Lovelace-powered NVIDIA RTX 4090 instances are ready for production use, hosted securely in the EU.
Our RTX 4090 24 GB (“PowerAI”) & RTX 4090 Pro 48 GB (“StellarAI”) instances offer:
We love AI to serve you with the best AI ever:



See how the RTX 4090 performs in some benchmarks compared to similar NVIDIA GPU types we also offer for renting.
| GPU | VRAM | TAIFlops | flux schnell | Llama 3.1 8B FP8 |
|---|---|---|---|---|
| RTX 4090 Pro | 48 GB | 238 | 2.63 s/image | 1221 tokens/s |
| RTX 4080 Super Pro | 32 GB | 139 | 19.9 s/image | 530 tokens/s |
| RTX 4090 | 24 GB | 133 | 12.6 s/image | 645 tokens/s |
| RTX 3090 | 24 GB | 100 | 19.3 s/image | n/a |
| V100 | 32 GB | 84 | 75.1 s/image | n/a |
| RTX 4070 Ti Super | 16 GB | 56 | n/a | 229 tokens/s |
Do you like the benchmark results? Try your workflow now:
Start RTX 4090 Server in Minutes
Explore our RTX 4090 server options and pricing. Servers hosted with a “.de” domain are located in Germany within ISO/IEC 27001 certified data centers.
24 GB VRAM:
| Blib Name | GPU | CPU RAM | Cores | NVMe | Price | +50% Promo Price** |
|---|---|---|---|---|---|---|
| powerai.s1 | 1× RTX 4090 24 GB | 32 GB | 4 | 180 GB | €0.41/h or €233/m | |
| powerai.m1 | 1× RTX 4090 24 GB | 58 GB | 8 | 600 GB | €0.45/h or €257/m | |
| powerai.l1 | 1× RTX 4090 24 GB | 88 GB | 12 | 900 GB | €0.48/h or €273/m | |
| powerai.m2 | 2× RTX 4090 24 GB | 82 GB | 16 | 900 GB | €0.83/h or €480/m | |
| powerai.l4 | 4× RTX 4090 24 GB | 256 GB | 36 | 2500 GB | €1.70/h or €977/m | |
| powerai.v4 | 4× RTX 4090 24 GB | 256 GB | 64 | 2500 GB | €1.78/h or €1023/m |
48 GB VRAM:
| Blib Name | GPU | CPU RAM | Cores | NVMe | Price | +50% Promo Price** |
|---|---|---|---|---|---|---|
| infinityai.m1 | 1× A100 40 GB | 78 GB | 8 | 900 GB | €0.71/h or €410/m | |
| infinityai.m1.de | 1× A100 40 GB | 78 GB | 8 | 900 GB | €0.76/h or €437/m | |
| infinityai.l1.de | 1× A100 40 GB | 92 GB | 20 | 1200 GB | €0.81/h or €467/m | |
| infinityai.m2 | 2× A100 40 GB | 140 GB | 10 | 600 GB | €1.30/h or €747/m | |
| infinityai.m2.de | 2× A100 40 GB | 140 GB | 10 | 1200 GB | €1.40/h or €803/m | |
| infinityai.m4 | 4× A100 40 GB | 356 GB | 30 | 2000 GB | €2.63/h or €1510/m | |
| infinityai.m4.de | 4× A100 40 GB | 356 GB | 30 | 2000 GB | €2.77/h or €1593/m | |
| infinityai.l6.de | 6× A100 40 GB | 512 GB | 120 | 3900 GB | €4.36/h or €2510/m | |
| infinityai.l8 | 8× A100 40 GB | 768 GB | 90 | 3900 GB | €5.29/h or €3043/m | |
| infinityai.l8.de | 8× A100 40 GB | 768 GB | 160 | 5350 GB | €5.83/h or €3353/m |
* Weekly and monthly subscriptions offer reduced rates. Promotional pricing includes a 50% extra credits bonus, valid for this month.
We meticulously maintain our servers, exemplified by builds like the RTX 4090 24GB & 48GB instances. This allows you to focus on your AI projects – development or production – without hardware concerns. As a boutique hyperscaler in Europe, we provide dedicated attention to detail and performance.
Rent your own GPU server today and start building amazing AI applications! Trooper.AI GPU servers are built from purely upcycled high-end tech from the last years, designed to provide you with the best performance, security, and reliability for all your AI needs.
EU location · High privacy · Great performance · Best support