MMOOZZGG.com

Hardware

AI server data center: where to put a server and what it costs per month

4 min read6 sources

A server with server-grade graphics cards is loud, runs hot and needs strong power, so an AI server data center is where it belongs. Below is the cost of hosting by our calculations, three scenarios from the office to your own data center, and what to check in the contract.

On this page · 4 sections

When a server needs a data center

A workstation with one card, such as an RTX 5090 with 32 GB, stays in the office: it draws about 800 W under load, and you hear it, but you can live with it. A server with two 96 GB cards draws up to 1.5 kW, and one with four H100 cards about 3.8 kW. Such a server with a UPS usually needs three-phase power, separate cooling and a place where the noise bothers nobody.

Server cards with blower fans are made for a rack, not for an office. So the rule is simple: one or two desktop cards live in the office, and a department server and anything bigger live in a data center.

How much hosting costs

Hosting in a commercial data center costs from $90 a month for a server up to 300 W, plus $24 for each additional 100 W. For a server with two RTX PRO 6000 cards at 1.5 kW this is about $400 a month, and for a server with four H100 cards at 4 kW about $1,000.

In our calculations, electricity is a separate line at a tariff of about $0.10 per kWh: a 1.5 kW server running around the clock uses about 1,080 kWh, or $100 a month. At some sites electricity is included in the price per kilowatt, so compare the total, not the line.

Office, colocation, rental or your own data center

Office: suits a Mac or a workstation with one card, with $0 for hosting. Commercial data center: the server is yours, and the rack and power belong to the site. By our calculation, a server with two RTX PRO 6000 cards costs about $2,100 a month, including depreciation, electricity, hosting, maintenance and an engineer.

GPU rental in a Russian data center: you don’t buy hardware, an H100 costs from $3 an hour, or from $1,700 a month at the cheapest offers. Counting the server, electricity and an engineer, buying pays off against renting in 18–54 months, by our calculation in the calculator.

You need your own data center when data cannot leave the company perimeter, not even for a Russian cloud or someone else’s rack: this is how a telecom operator, an airport or state bodies set the task. Department and division servers don’t need their own data center.

What to check in the contract

A site in Russia and a contract that shows who is responsible for the data: this matters for personal data under 152-FZ, Russia’s personal data law. Power per rack for GPUs in kilowatts, not only space in units. Whether electricity is included in the price, and how overuse is billed.

Your engineer’s access to the server, the procedure for replacing a card and the time the site needs to install it. A GPU rental contract signed in advance as a backup: switching it on in an hour is cheaper than a week of downtime.

Questions and answers

Do you need your own data center to run AI locally?

No. A Mac or a workstation stays in the office, and a department server goes to a commercial data center from $90 a month for a server up to 300 W. You need your own data center only if data cannot leave the company perimeter.

Which is cheaper: hosting your own server or renting GPUs?

Rental pays off for a pilot, seasonal load and backup. Your own server pays off under steady round-the-clock load that lasts longer than the payback period, which by our calculation is from one and a half to four and a half years.

Can you put a server with four cards in an office?

Technically yes, but 3.8 kW usually needs three-phase power, a UPS and cooling, and the noise gets in the way of work. It is simpler and often cheaper to host such a server in a data center.

Sources and fact-checking

Checked by the editors on 16 October 2026.

  1. HOSTKEY, colocation in Moscow, 2026 (hostkey.ru)
  2. RusCable, electricity tariffs in Moscow from 1 January 2026 (ruscable.ru)
  3. NVIDIA, RTX PRO 6000 Blackwell: 96 GB, 600 W (nvidia.com)
  4. NVIDIA, H100: SXM 80 GB up to 700 W, NVL 94 GB 350–400 W (nvidia.com)
  5. GPU Index, H100 rental in Russia: 41 offers, from $3 an hour (checked 30 September 2026) (gpuindex.ru)
  6. Federal Law No. 152-FZ “On Personal Data”, Russia’s personal data law, text on ConsultantPlus (consultant.ru)

Found a mistake? Tell us

Publications

Cases on this topic: 8

  • Market case · Russia

    A large Russian bank in-house LLMs solve 45% of requests; support costs estimated down at least 25% a year

    The bank put its own family of language models into its chatbot for LLM customer support: about 45% of customer requests are solved without an operator. In October 2025 it launched AI copilots for first-line operators, and more than 90% of support staff use them every day. The models run inside the bank’s network perimeter, so data does not leave the bank.

    Banks and finance

    source: platforms.su

    Read the case
  • Market case · Russia

    A large Russian financial group AI financial assistant on in-house language models, beta since 20 July 2026

    The group built an AI financial assistant for tasks linked to money and shopping, on its own family of large language models. To avoid made-up answers, the system relies on verified data sources, not only on what the model has learned. The beta was presented on 20 July 2026, and the full launch is planned for the autumn.

    Banks and finance

    2 September 2026 · source: platforms.su

    Read the case
  • Market case · Russia

    Alfa-Bank, IT head local agent on the Qwen 3.6–35B-A3B model: about 70 tokens per second on a Mac

    A local AI agent on Mac does autonomous work on a model with 35 billion parameters, about 3 billion of them active per token. It runs through LM Studio at about 70 tokens per second. The agent first ran in Docker on the Mac, then moved to a virtual machine, and the author calls that move a turning point for ease of use.

    Banks and finance

    2 September 2026 · source: platforms.su

    Read the case
  • Market case · Russia

    Ministry of Justice of Russia AI lawyer avatar with a base of federal laws; a local language model is being trained for it

    The country’s first AI lawyer avatar understands spoken language and advises on legal questions; its knowledge base holds federal laws, regulations and information from the ministry’s website. As of July 2026, specialists are training a local language model, which runs on the organization’s own servers, to speed up the work and improve the answers.

    Government

    2 September 2026 · source: platforms.su

    Read the case
  • Market case · Russia

    Moscow City Health Department AI assistant for hospital analytics on a local language model: data does not leave the network

    An AI assistant built on a locally deployed large language model sits inside the data marts of the city’s unified medical system. Local deployment was chosen to work with sensitive medical data without sending it outside.

    Government

    2 September 2026 · source: platforms.su

    Read the case
  • Market case · Russia

    Pulkovo Airport AI secure code review plugin in a closed network: over 90% accuracy on checking findings and up to 85% on fixes

    The airport’s information security team built code analysis into the development environment. A trained AI plugin reviews vulnerabilities and fixes the code inside a closed network, so no data goes outside. The model verifies findings with more than 90% accuracy and fixes vulnerabilities with up to 85% accuracy.

    Industry and infrastructure

    2 September 2026 · source: platforms.su

    Read the case
  • Market case · Russia

    Step agroholding its own AI ecosystem on local neural networks, so data never goes to public services

    The holding builds its AI ecosystem on local AI models for companies, agents and chatbots, so confidential data does not go to public services. It already syncs herd management data with a government system and runs legal information monitoring for the back office. Several analytics projects have started.

    Industry and infrastructure

    2 September 2026 · source: platforms.su

    Read the case
  • Market case · Russia

    Developer MR Project finance AI controller in an isolated environment: 80% less time on credit document analysis

    The developer and its bank built a project finance AI controller on a Russian language model. It checks the terms of project finance automatically and tracks loan obligations. According to the company, the time spent on credit document analysis fell by 80%. Each large project comes with hundreds of pages of credit documents and dozens of additional agreements; the system assembles a master version of each contract and keeps one calendar of obligations.

    Industry and infrastructure

    15 September 2026 · source: comnews.ru

    Read the case
Contact us