Best GPU for running local LLMs
Pick a model family. Each guide shows the cheapest GPU and the cheapest Mac or mini-PC that run every model in it.
Guides
- Best GPU for running Llama locally3 models
- Best GPU for running Qwen3 locally4 models
- Best GPU for running gpt-oss locally2 models
- Best GPU for running DeepSeek R1 locally4 models
- Best GPU for running Gemma 3 locally3 models
- Best GPU for running Gemma 4 locally5 models
- Best GPU for running Qwen3.8 locally3 models
- Best GPU for running Ornith 1.5 locally3 models
- Best GPU for running Nemotron locally3 models
- Best GPU for running Liquid AI LFM locally2 models
- Best GPU for running Mistral locally4 models
- Best GPU for running GLM locally4 models
- Best GPU for running Phi locally4 models
- Best GPU for running long-memory models locally3 models
- Best GPU for running local coding models locally8 models
By memory size
- Best LLMs for 4 GB VRAM4 GB
- Best LLMs for 6 GB VRAM6 GB
- Best LLMs for 8 GB VRAM8 GB
- Best LLMs for 10 GB VRAM10 GB
- Best LLMs for 12 GB VRAM12 GB
- Best LLMs for 16 GB VRAM16 GB
- Best LLMs for 20 GB VRAM20 GB
- Best LLMs for 24 GB VRAM24 GB
- Best LLMs for 32 GB VRAM32 GB
- Best LLMs for 48 GB VRAM48 GB
- Best LLMs for 64 GB VRAM64 GB
- Best LLMs for 96 GB VRAM96 GB
- Best LLMs for 128 GB VRAM128 GB
By budget
- Best GPU for local AI under $3008 devices
- Best GPU for local AI under $50025 devices
- Best GPU for local AI under $80044 devices
- Best GPU for local AI under $1,00053 devices
- Best GPU for local AI under $1,50059 devices
- Best GPU for local AI under $2,00063 devices
- Best GPU for local AI under $3,50068 devices
By model size
How much VRAM do you need? · Ollama VRAM requirements by tag · GGUF VRAM requirements by quant · 1B–2B · 2B–3B · 4B–6B · 7B–8B · 12B–14B · 16B–21B · 24B–27B · 30B–32B · 40B–50B · 70B · 120B · 200B–300B · 300B–480B · 500B–800B · 1T+
Overall picks
Best GPU for running local LLMs · Best Mac for running local LLMs · Best mini PC for local LLMs · Best used GPU for local LLMs · Best AMD GPU for local LLMs · Best Intel GPU for local LLMs · NVIDIA vs AMD vs Mac · Tokens per second by GPU and Mac · Multi-GPU for local LLMs · Q4 vs Q8 vs FP16 · Context length and VRAM · MoE vs dense models · GPU power consumption
Head-to-head
GPU vs GPU comparisons for local LLMs · Model vs model (VRAM and speed)