From model
to machine.
Accelerators, complete systems and the infrastructure around them. Understand the trade-offs. Specify the right build.
Hardware in context.
| Equipment | Class | Key specification | Read / shortlist |
|---|---|---|---|
| NVIDIANVIDIA A100 | Accelerator | 40 / 80 GB · PCIe or SXM | |
| NVIDIANVIDIA H100 | Accelerator | 80 GB SXM / 94 GB NVL | |
| AppleMac Studio M3 Ultra (2025) | AI computer | 2025 platform · configuration-specific RAM | |
| NVIDIANVIDIA DGX Spark | AI computer | 128 GB unified memory |
Make the comparison.
View comparisons →A100 vs H100: Choose by Workload and Complete System Cost
The useful comparison is between two complete, runnable configurations. Match the memory, form factor, software and commercial scope before deciding whether the H100 premium is justified.
17 Nov 2026 Read →ComparisonDGX Spark vs Mac Studio for Local AI
Compare a specific DGX Spark configuration with a specific M3 Ultra Mac Studio. Start with the software you must run, then test whether the available memory and sustained performance meet the same workload.
17 Nov 2026 Read →Servers, storage and everything between.
A CPU choice, a drive label and a RAID card do not define a working system. See how the pieces connect.
Models meet hardware.
All model profiles →gpt-oss-20b: Local Hardware and Deployment Requirements
Start with the official artifact and a runtime that supports its format. The publisher's memory example is a useful starting point, but the production requirement also depends on context, execution settings and concurrent requests.
17 Nov 2026 Read →AI modelgpt-oss-120b: GPU Memory and System Selection
The model's parameter count does not directly translate into a single universal memory requirement. Use the intended artifact and execution path, then validate the complete request profile on candidate hardware.
17 Nov 2026 Read →AI modelQwen3.5-35B-A3B: Hardware Sizing and Multimodal Requirements
Size this model from the full artifact and supported execution path. The active-parameter count is a compute characteristic, not a shortcut for the total memory needed to host the model.
17 Nov 2026 Read →