GPU servers with NVIDIA RTX PRO 6000

Servers with NVIDIA RTX PRO 6000 (Blackwell, 96 GB): cloud, VPS and dedicated. Compare GPU count, VRAM, CPU and price; cheapest first.

Filters 1
Processor
Disk
Disk type
Size, GB
RAM
Traffic per month
at least
Operating system
Country
Payment method
Virtualization
Terms
Found 14 plans — showing , first:
Host and plan ProcessorRAM Disk Traffic Where Payment Price
MojoHost MojoHostMOJOGPU RTX PRO 6000 Max Q 24GB
GPU NVIDIA RTX PRO 6000 Max Q, 24 GB Hourly billing DDoS protection Ready in minutes
14 coresCPU 60 GBRAM —disk —traffic USA Bank card (Visa/Mastercard)
249 $ Open
Hukot HukotGPU-VPS5
GPU NVIDIA RTX PRO 6000, 96 GB Ready in minutes IPv6
20 coresCPU 60 GBRAM 1 TBSSD unlimitedtraffic Czech Republic Bank card (Visa/Mastercard)PayPalInvoice for companies
756 $ when paying for year: 703 $/⁠mo Open
Microsoft Azure Microsoft AzureNC24ldsxlRTX6K v6
GPU NVIDIA RTX PRO 6000 Blackwell Server Edition Trial Hourly billing Ready in minutes
24 coresCPU 72 GBRAM —disk —traffic Bank card (Visa/Mastercard)
825 $ when paying for 3 years: 363 $/⁠mo Open
Alpenhost e.U. Alpenhost e.U.GPU-vServer Platin
GPU NVIDIA RTX Pro 6000 Blackwell, 96 GB DDoS protection Ready in minutes IPv6
16 coresCPU 64 GBRAM —disk 1 Gbit/sport Austria PayPalInvoice for companies
879 $ Open
LeaseWeb LeaseWebHP DL385 G11 (8SFF) 2x AMD EPYC 9224 - RTX6000 (MTL-02)
GPU NVIDIA RTX PRO 6000, 96 GB Trial
48 cores2.5 GHz 128 GBRAM 1920 GBSSD 30 TB1 Gbit/s Canada Bank card (Visa/Mastercard)PayPalInvoice for companies
1,074 $ when paying for 3 years: 806 $/⁠mo Open
Hetzner HetznerGEX63
GPU NVIDIA RTX PRO 6000 Blackwell Max-Q, 96 GB Hourly billing
20 cores2.4 GHz 128 GBRAM 3840 GBNVMe unlimited1 Gbit/s Finland Bank card (Visa/Mastercard)PayPalInvoice for companies
1,098 $ Open
Contabo.com Contabo.comGPU VPS
GPU NVIDIA RTX PRO 6000 Blackwell Server Edition, 96 GB DDoS protection Ready in minutes IPv6
18 coresCPU 96 GBRAM 900 GBNVMe unlimited1 Gbit/s Germany Bank card (Visa/Mastercard)PayPalInvoice for companies
1,098 $ Open
Infinitive Host Infinitive HostGPU Surge
GPU NVIDIA RTX PRO 6000, 96 GB Money-back
24 coresCPU 256 GBRAM 960 GBNVMe unlimited1 Gbit/s Germany Bank card (Visa/Mastercard)Invoice for companies
1,103 $ Open
Zeno Cloud Zeno CloudNVIDIA RTX PRO 6000 96GB, 16 vCPU / 128 GB
GPU NVIDIA RTX PRO 6000, 96 GB
16 coresCPU 128 GBRAM —disk —traffic India Bank card (Visa/Mastercard)
1,178 $ Open
Cloudzy.com Cloudzy.comGPU VPS 1× RTX PRO 6000 Blackwell
GPU NVIDIA RTX PRO 6000 Blackwell, 96 GB Hourly billing Money-back DDoS protection
24 coresCPU 200 GBRAM 1300 GBNVMe —traffic CryptocurrencyBank card (Visa/Mastercard)PayPalInvoice for companies
1,300 $ yearly: 1,300 $/⁠mo Open
Akamai (Linode) Akamai (Linode)RTX PRO 6000 Blackwell x1
GPU NVIDIA RTX PRO 6000 Blackwell Server Edition, 96 GB Hourly billing IPv6
16 coresCPU 176 GBRAM 1024 GBSSD unlimited16 Gbit/s Bank card (Visa/Mastercard)PayPalInvoice for companies
2,190 $ Open
Beget BegetBlackwell №2
GPU NVIDIA Quadro RTX PRO 6000, 96 GB Trial Hourly billing Ready in minutes
48 cores2.8 GHz 128 GBRAM 4 TBNVMe 30 TB1 Gbit/s Russia Invoice for companiesYooMoney
3,186 $ Open
Timeweb Cloud Timeweb CloudAMD Ryzen 9 9950X (16 ядер, 4.3-5.7 ГГц, 32 потока) / 192 DDR5 / 2 x 2 Тб NVMe + GPU RTX PRO 6000 96GB
GPU NVIDIA RTX PRO 6000, 96 GB Hourly billing Ready in minutes
16 cores4.3 GHz 192 GBRAM 4 TBNVMe —traffic Russia Bank card (Visa/Mastercard)Invoice for companiesYooMoney
3,284 $ Open
Timeweb TimewebAMD Ryzen 9 9950X (16 ядер, 4.3-5.7 ГГц, 32 потока) / 192 DDR5 / 2 x 2 Тб NVMe + GPU RTX PRO 6000 96GB
GPU NVIDIA RTX PRO 6000, 96 GB Trial Hourly billing
16 cores4.3 GHz 192 GBRAM 4 TBNVMe unlimited1 Gbit/s Russia Bank card (Visa/Mastercard)Invoice for companiesYooMoney
3,284 $ Open

GPU servers with NVIDIA RTX PRO 6000: prices, hosting providers and payment methods

The selection has 50 plans from 14 hosts. Prices start at 249 $ per month, a quarter of plans are cheaper than 1,123 $, half are cheaper than 1,505 $. NVMe: 24% of plans, unlimited traffic: 20%, hourly billing: 62%.

Cheapest hosting providers

  1. MojoHostMOJOGPU RTX PRO 6000 Max Q 24GB · 14 cores · 60 GB249 $
  2. HukotGPU-VPS5 · 20 cores · 60 GB756 $
  3. Microsoft AzureNC24ldsxlRTX6K v6 · 24 cores · 72 GB825 $
  4. Alpenhost e.U.GPU-vServer Platin · 16 cores · 64 GB879 $
  5. LeaseWebHP DL385 G11 (8SFF) 2x AMD EPYC 9224 - RTX6000 (MTL-02) · 48 cores · 128 GB1,074 $

Price by RAM size

RAMPlansPrice from
from 1 GB50249 $
from 2 GB50249 $
from 4 GB50249 $
from 8 GB50249 $
from 16 GB50249 $

Payment methods

  • Bank card12 86%
  • Cryptocurrency1 7%

How many hosting providers in the selection accept each payment method.

Where most plans are

Provider comparisons VPS price index VPS price calculator VPS guide

RTX PRO 6000 for rent: 96 GB on a single card for large models and rendering

The NVIDIA RTX PRO 6000 is a professional Blackwell-generation card with 96 GB of GDDR7 memory with error correction (ECC). That is more than the 80 GB of an H100 PCIe and twice as much as an RTX 6000 Ada. The main reason to pick a GPU server with RTX PRO 6000 is to run a model on one card that used to need two, without splitting the weights across accelerators. It is still a full graphics card, too: it has RT cores for ray tracing and hardware NVENC encoders, so people rent it for rendering and video as well.

Three cards with similar names

Catalogs make it easy to mix up three different generations, and their prices differ several times over:

  • Quadro RTX 6000 — Turing architecture, 24 GB. An old graphics card with too little memory and compute for current neural networks.
  • RTX 6000 Ada — 48 GB, the previous generation. Close in design to the L40S.
  • RTX PRO 6000 Blackwell — 96 GB of GDDR7, fifth-generation tensor cores with FP4, FP6 and FP8 formats. This is the card collected on this page.

Blackwell comes in variants as well. The Server Edition is built for racks: passive cooling, up to 600 W, with airflow from the server fans. Max-Q is a 300 W workstation version with a blower fan; it is easier to fit several in one chassis, but under sustained load it runs at lower clocks. If the plan line just says "RTX PRO 6000", ask the host which variant it is: speed depends on it, and so does MIG availability.

What fits into 96 GB

A rough estimate: in FP16 a model takes about 2 GB per billion parameters, with 8-bit quantization about 1 GB, with 4-bit 0.5–0.6 GB. On top of that you need headroom for the KV cache, which grows with context length and the number of concurrent requests.

  • FP16 without quantization — models up to roughly 30–40 billion parameters with moderate context.
  • 8-bit — a 70-billion-parameter model entirely on one card, with room left for the cache.
  • 4-bit — models in the 100–120 billion class. FP4 support in the tensor cores speeds up this kind of inference if your engine handles the format.
  • Fine-tuning — LoRA for 13–30B models in 16-bit and QLoRA for 70B. Full training of large models is not a job for this card.
  • Images and video — image generation in large batches, heavy video generation models that feel cramped in 24–48 GB.
  • Rendering and 3D — scenes with huge textures that do not fit in gaming cards, offline rendering, video transcoding on NVENC.

MIG, multiple cards and when an H100 is better

The Server Edition supports MIG: the card can be split into four isolated instances of about 24 GB each, every one with its own memory and compute units. That is handy when several small models or several users share one server. If a plan lists a share of the card with part of its memory, it is a vGPU or MIG slice, not a whole card: compare the memory size, not the model name.

The "×2" and "×4" plans give 192 and 384 GB in total, but the RTX PRO 6000 has no NVLink, so the cards talk over PCIe 5.0. For inference this usually does not get in the way: each card can run its own copy of the model, and splitting one model across two cards over PCIe works acceptably. For training, where cards constantly synchronize gradients, the slower link eats up the gain. If you plan multi-GPU training, look at H100 servers with NVLink.

Driver, CUDA and the image

The card has compute capability 12.0. It needs the R570 driver branch or newer and CUDA 12.8 or later. In older images with CUDA 12.4 the card either will not be detected or the framework will find no kernels for it. Once the server is delivered, check the nvidia-smi output, the CUDA version inside the container, and that PyTorch or your inference engine is built with support for this architecture.

Hourly billing or a month

Renting a server with RTX PRO 6000 in the catalog costs roughly $830 to $5,700 a month, with a typical plan around $1,800. The low end is a single cloud card, the high end is dedicated machines with four cards. Most plans are billed by the hour, which pays off for a trial run, a one-off data processing job or a render against a deadline: you pay only for the hours the machine runs. If the card will be busy more than two thirds of the month, compare with the monthly price and with the dedicated GPU servers section. Before paying, check that the host accepts your card in your currency; if you process personal data of EU residents, also check the data centre country for GDPR purposes and its latency to your users.

FAQ

10 questions with answers

How much video memory does the RTX PRO 6000 have?

96 GB of GDDR7 with ECC error correction. That is more than the H100 PCIe (80 GB) and twice the RTX 6000 Ada.

Will a 70-billion-parameter model fit on one RTX PRO 6000?

With 8-bit or 4-bit quantization, yes, with room for context. In FP16 such a model takes about 140 GB and needs two cards.

How is the RTX PRO 6000 different from the RTX 6000 Ada and Quadro RTX 6000?

They are three generations: Quadro RTX 6000 (Turing, 24 GB), RTX 6000 Ada (48 GB) and RTX PRO 6000 Blackwell (96 GB, FP4 support). For neural networks the gap in memory and speed is very large.

Does the RTX PRO 6000 have NVLink?

No, in multi-GPU servers the cards are linked over PCIe 5.0. That is usually enough for inference; for training across several cards H100s with NVLink are a better fit.

Which CUDA version does the RTX PRO 6000 need?

CUDA 12.8 or newer and a driver from the R570 branch or later. In images with an older CUDA the framework may not recognise the card.

What does Max-Q mean in a plan name?

It is a 300 W variant of the card instead of the 600 W Server Edition. It has the same 96 GB of memory, but runs slower under sustained load.

How much does GPU servers with NVIDIA RTX PRO 6000 cost?

From 249 $ per month. A quarter of plans are cheaper than 1,123 $, half are cheaper than 1,505 $. In total 50 plans from 14 hosts.

Can I pay for GPU servers with NVIDIA RTX PRO 6000 by bank card or cryptocurrency?

By bank card at 12 hosts of 14 (86%), by cryptocurrency at 1 (7%). Payment methods for each plan are in the Payment column.

How much does GPU servers with NVIDIA RTX PRO 6000 with 2 GB of RAM cost?

From 249 $ per month; with 4 GB from 249 $, with 8 GB from 249 $.

Which countries have the most plans?

Most plans are in in the USA, in Germany and in the UK. You can pick a country in the filter, and prices and payment methods will update.

Related

Guide
GPU Server: Which Graphics Card for Which Job — Memory, Bandwidth and PricesYour Own Language Model on a Server: How Much Video Memory and Which GPU You Need
Comparisons
Hetzner vs Contabo.comAkamai (Linode) vs Microsoft AzureTimeweb Cloud vs Beget

Prices and specs are taken from hosting providers' websites; check them on the website before paying, as hosts change plans without notice. About the catalog and methodology