How to choose a computer for local LLMs
Last updated
Choose a computer for local LLMs by the largest model you need, then by memory bandwidth, software support, single or shared use, and price. Olares One suits shared use up to about 32B on CUDA; larger models need more memory.
Olares One → Computer for local LLMs → Ollama · vLLM · LM Studio → Local models without the cloud
Criteria
- Memory available to the model, which sets the largest model that fits
- Memory bandwidth, which sets how fast tokens come out
- Software support: CUDA, Apple Metal or AMD
- One user or a shared server for several people and devices
- Complete price, including any parts, OS and setup time
- Power draw, size and noise where the computer will sit
Machines compared
| Machine | Price | Memory for models | Memory type | GPU | CUDA |
|---|---|---|---|---|---|
| Olares One | $4,999 · €4,599.95 | 24 GB | VRAM | RTX 5090 Mobile | yes |
| NVIDIA DGX Spark | $6,950 | 128 GB | Unified, 273 GB/s | NVIDIA GB10 | yes |
| ASUS Ascent GX10 | $5,999 | 128 GB | Unified | NVIDIA GB10 | yes |
| Apple Mac Studio (M5 Ultra) | $5,499 · €6,599 | 96 GB | Unified, 1.2 TB/s | M5 Ultra, 64-core GPU | no |
| Apple Mac Studio (M5 Max) | $2,499 · €2,999 | 36 GB | Unified, 460 GB/s | M5 Max, 32-core GPU | no |
| Apple Mac mini (M5 Pro, 64GB) | $3,199 · €3,649 | 64 GB | Unified, 307 GB/s | M5 Pro, 20-core GPU | no |
| Minisforum MS-S1 Max | $3,959 | 128 GB | Shared (unified) | Radeon 8060S | no |
| GMKtec EVO-X2 (Ryzen AI Max+ 395, 128GB) | $3,499.99 | 128 GB | Shared (unified) | Radeon 8060S | no |
| Framework Desktop (Ryzen AI Max+ 395, 128GB) | $3,449 | 128 GB | Shared (unified) | Radeon 8060S | no |
| Beelink GTR9 Pro (Ryzen AI Max+ 395, 128GB) | $4,199 | 128 GB | Shared (unified) | Radeon 8060S | no |
| Used RTX 3090 tower build | $3,800 | 24 GB | GDDR6X VRAM, 936 GB/s | RTX 3090 (used) | yes |
| NVIDIA GeForce RTX 5090 (desktop card) | $1,999 | 32 GB | GDDR7 VRAM | RTX 5090 desktop | yes |
Prices and specs read on the makers' pages on 12-13 November 2026; each name links to its source.
Should you pick the model or the machine first?
Pick the model first. Write down the largest model you will run and add room for context. A 32B model needs about 19-20 GB, a 70B model about 40 GB and a 120B model about 65 GB (approximate rule of thumb at 4-bit). That one number rules machines in or out before price, brand or speed enter the decision.
Then decide between one user and many. One user can run LM Studio or Ollama on a desktop. A shared server needs accounts, remote access and an engine that handles several requests at once.
How do the main options score?
Each option wins on a different criterion. Olares One leads on shared use with CUDA and trails on capacity. The Mac Studio M5 Ultra and DGX Spark lead on capacity. The used RTX 3090 tower leads on price for 24 GB with CUDA. The Minisforum MS-S1 Max gives 128 GB of shared memory without CUDA.
| Machine | Model memory | CUDA | Price (read 12-13 Nov 2026) |
|---|---|---|---|
| Olares One | 24 GB VRAM + 96 GB RAM | Yes | 4,999 USD |
| Used RTX 3090 tower | 24 GB VRAM | Yes | about 3,800 USD (SpecPicks estimate, no case) |
| Mac Studio M5 Ultra | 96 GB unified (base) | No | 5,499 USD |
| NVIDIA DGX Spark | 128 GB unified | Yes | 6,950 USD |
| Minisforum MS-S1 Max | 128 GB shared | No | 3,959 USD (sale) |
See the best PC and under 5,000 USD for more.
Questions & answers
Should I buy more memory than I need today?
Some headroom helps, because longer context and newer models use more. Do not pay for 128 GB if every model you use fits in 24 GB.
Does the CPU matter for local LLMs?
Less than the GPU or memory. It matters when a model spills out of VRAM into system RAM, where CPU and RAM speed set the pace.
Sources for this entry
- Olares One specifications (Olares docs) (Company-published)
- NVIDIA DGX Spark: maker's page (Maker-published)
- ASUS Ascent GX10: maker's page (Maker-published)
- Apple Mac Studio (M5 Ultra): maker's page (Maker-published)
- Apple Mac mini (M5 Pro, 64GB): maker's page (Maker-published)
- Minisforum MS-S1 Max: maker's page (Maker-published)
- GMKtec EVO-X2 (Ryzen AI Max+ 395, 128GB): maker's page (Maker-published)
- Framework Desktop (Ryzen AI Max+ 395, 128GB): maker's page (Maker-published)
- Beelink GTR9 Pro (Ryzen AI Max+ 395, 128GB): maker's page (Maker-published)
- Used RTX 3090 tower build: maker's page (Maker-published)
- NVIDIA GeForce RTX 5090 (desktop card): maker's page (Maker-published)
Continue this reading path
How the related websites connect
Cite this entry
Best Computer for Local LLM. “How to choose a computer for local LLMs.” Content reviewed 13 November 2026.
https://bestcomputerforlocalllm.com/choosing/