Best home rig for DeepSeek R1 Distill 14B

Older model. For current picks see 2025–2026 models.

Real, measured generation speeds for DeepSeek R1 Distill 14B on home hardware, fastest first. Different quants/engines aren't directly comparable.

Build	Engine	tok/s
Intel Core Ultra 265K + RTX 4090	llama.cpp	92.49	source
Intel Core Ultra 265K + RTX 3080 Ti	llama.cpp	70.74	source
Intel Core Ultra 265K + Radeon AI Pro R9700	llama.cpp	53.72	source
Mac Studio M3 Ultra 512GB (Geerling test)	llama.cpp	51.85	source
Intel Core Ultra 265K + RTX 4070 Ti	llama.cpp	49.74	source
Intel Core Ultra 265K + RTX 2080 Ti	llama.cpp	45.06	source
Mac Studio M1 Ultra (64-core GPU) 128GB	llama.cpp	41.51	source
Intel Core Ultra 265K + RTX 3060	llama.cpp	29.77	source
Dell Pro Max with GB10 (Nvidia DGX Spark)	llama.cpp	23.92	source

SPONSOR
Your brand here. One sponsor slot on this page, shown to people choosing AI hardware. Sponsor RigBase →

Rent before you buy: try this model on a cloud GPU for a few dollars first: RunPod

FAQ

RTX 4090: 92.49 tok/s generation (llama.cpp), from https://github.com/geerlingguy/ai-benchmarks/issues/43.