What server for AI? Guide and pricing by tier (2026)
The price of a server for AI depends on the tier, the number of cards, and the memory. In 2026, the memory shortage is pushing everything upward. Here are indicative ranges by segment.
What sets the price of an AI server
Three factors dominate: the number and type of graphics cards, the amount of memory, and the server form factor. Memory weighs more than ever in the total. The full calculation, hardware versus pay-per-token, is in our article on-premises versus API cost.
Memory, the driver of 2026 prices
Manufacturers have redirected their capacity toward high-bandwidth memory for AI accelerators, which yields more revenue per wafer than standard memory. The result is a scarcity of ordinary system memory. Over one year, server memory prices have risen sharply, and vendors have raised their prices by fifteen to twenty percent. A workstation such as the DGX Spark saw its price rise from around four thousand to nearly four thousand seven hundred dollars, purely because of memory cost. Relief is not expected for several years.
Price ranges by tier
The figures below give order-of-magnitude estimates for 2026, taxes and integration excluded. They are deliberately wide, since configuration and memory can double the total.
| Tier | Who it's for | Indicative range |
|---|---|---|
| DGX Spark workstation (128 GB) | Small business, sovereign workstation | 4,500 to 6,000 € |
| Mac Studio Ultra (512 GB) | Small business, silent option | 30,000 to 45,000 € |
| DGX Station workstation (GB300) | SME without a server room | 90,000 to 130,000 € |
| RTX PRO 6000 server (2 to 8 GPUs) | SME, workhorse | 250,000 to 600,000 € |
| H200 server (8 SXM GPUs) | Large enterprise | 400,000 to 900,000 € |
| GB300 NVL72 rack | Sovereign AI cloud | from 3 M€ |
Indicative 2026 estimates, excluding discounts, hosting, and integration. Precise pricing depends on the configuration chosen.
Hardware partners
The calculation relies on NVIDIA graphics cards. Servers are certified by HPE, Dell, Supermicro, Lenovo, ASUS, and Gigabyte, which brings the vendor support and warranty expected in enterprise settings. Dell and Lenovo notably raised their server prices as soon as the shortage began. QDNA integrates and operates this hardware, from workstation to rack. Detailed tiers are in our hardware sheets.
Buy, colocate, or rent
Three paths coexist. Buying ties up capital but amortizes the cost over several years, with a cost per token close to zero. Managed colocation in a sovereign French data center avoids buying premises and keeps legal control. Renting capacity suits peaks or validation, ahead of an investment. The cost break-even point is detailed in the article on-premises versus API cost.
Sizing correctly before buying
Model size and number of users set the tier. A compact model fits on a workstation, a large model requires a server. The practical guide is in deploying a local LLM. In a tight memory market, correct sizing avoids overpaying for an oversized configuration.
Comparing the real cost per token
The purchase price does not tell the whole story. What matters is the cost per token, which depends on hardware throughput for a given model. The open comparator InferenceX by SemiAnalysis publishes reproducible, auditable measurements: latency, throughput in tokens per second, and implied cost per token, for many open-weight models (DeepSeek, Kimi, GLM, MiniMax, Qwen) and across the leading accelerators, from NVIDIA Hopper and Blackwell to AMD cards. It helps choose the hardware with the best performance-to-price ratio for a specific workload, rather than relying on the spec sheet alone.
Frequently asked questions
How much does an AI server cost in 2026?
From a workstation at a few thousand euros to a rack at several million, depending on the tier. An SME server with several cards sits between several tens and several hundreds of thousands of euros depending on memory. Figures are indicative and pushed up by memory.
Why are AI servers so expensive in 2026?
Manufacturers have redirected capacity toward high-bandwidth memory for AI, which has made standard memory scarcer. Server memory prices have risen sharply and several vendors have raised their prices.
What AI server for an SME?
A server fitted with RTX PRO 6000 cards, from two to eight depending on need, covers most SME use cases. The indicative range runs from around forty-five thousand to one hundred and ninety thousand euros depending on configuration.
Should you buy or rent an AI server?
Buying amortizes the cost over several years with a low cost per token. Renting suits peaks and validation. Sovereign colocation avoids buying premises while keeping legal control.
What is the total cost of ownership of an AI server?
The purchase price is only part of the cost. You must add electricity, which depends on card power draw and cooling, depreciation over three to five years, maintenance, and hosting. Over time, an amortized server keeps a cost per token well below that of a usage-billed API.
Let's size your AI server
Sizing and an estimate based on your actual usage, purchase or sovereign colocation.
Book a call