Hardware for Local LLMs
Specs for 26 GPUs and Apple Silicon devices stay on this page. Separate per-SKU URLs were consolidated here for good — they return 410 and are not coming back.
All 26 GPUs, mini PCs, and Apple Silicon devices tracked for local AI inference. Use the VRAM calculator to see which models fit each device, or open the compare table. Kept guides such as the RTX 4090 LLM guide stay on their own URLs.
NVIDIA GPUs
AMD GPUs
Intel GPUs
Apple Silicon
In stock (2026-10-01)
In stock (2026-10-01)
Price needs re-check (no clean buy box, 2026-10-01)
Price needs re-check (no clean buy box, 2026-10-01)
In stock — sold by Amazon.com (2026-10-01)
Currently unavailable (2026-10-01)
Currently unavailable (2026-10-01)
Price not re-checked (2026-10-01) — configure 64GB+ at Apple
Mini PCs
96–128 GB unified at ~256 GB/s. Q4: dense 70B fits and is slow (~5 tok/s class); Qwen3.8-Flash-Next MoE fits; DeepSeek-V4-Flash only at an aggressive sub-Q4 quant (~155 GB at a normal 4-bit). Not CUDA.
In stock — sold by GMKtec-US ($3,649.99, 2026-10-02)
Same Q4 fit as the EVO-X2: 128 GB at ~256 GB/s. DeepSeek-V4-Flash only aggressive; Qwen3.8-Flash-Next MoE; dense 70B Q4 slow. Not the default Halo buy.
Low stock — only 3 left, Beelink Direct, $4,349.00 on 2026-10-01. Confirm NIC revision before buying.
Same 96–128 GB at ~256 GB/s tier when stocked. Not a buy: unavailable on 2026-10-01, 3.4 stars from 20 product-specific ratings, repeated 10GbE failures.
Out of stock on 2026-10-01. No price. 3.4/20 product-specific ratings, NIC failures.