Hardware

8GB GPUs for local AI

The entry tier. Runs 4B-9B models fully on the GPU at Q4 quantisation, which covers everyday chat, summarising and light coding.

Models that run well on this tier

gemma4:e4b

E4B (effective 4B) · Q4_K_M · ~8.1 GB

Ollama's default gemma4 tag. Good 8GB pick.

qwen3.5:9b

9B · Q4_K_M · ~8.1 GB

Excellent 8-12GB all-rounder.

ministral-3:8b

8B · Q4_K_M · ~7.5 GB

Strong 8GB Mistral model.

qwen3:8b

8B · Q4_K_M · ~6.7 GB

Solid 8GB reasoning model.

deepseek-r1:8b

8B · Q4_K_M · ~6.7 GB

Shows its chain of thought; good maths on 8GB.

qwen2.5-coder:7b

7B · Q4_K_M · ~6.2 GB

Reliable small coder and autocomplete model.

Check your exact setup in the GPU picker → · Read the full guide →

Hardware picks

NVIDIA GeForce RTX 5060 (8GB)

8GB GDDR7 · bandwidth 448 GB/s

Current-gen entry card with fast GDDR7. Fine for 8B-class models; you'll outgrow it fast if you want 14B+.

Check price

NVIDIA GeForce RTX 4060 (8GB)

8GB GDDR6 · bandwidth 272 GB/s

Low power and widely available used. Slower memory means fewer tokens per second than the 5060.

Check price

Intel Arc B580 (12GB) — the budget upgrade

12GB GDDR6 · bandwidth 456 GB/s

Often priced like an 8GB card but gives you 12GB. Works with Ollama through Vulkan; expect more setup friction than NVIDIA.

Check price
Our advice: If you're buying new specifically for local AI, skip 8GB and stretch to 12-16GB. VRAM is the single biggest factor in which models you can run.

We don't list prices because they change daily. Bandwidth figures are manufacturer specs. Disclosure: "Check price" links may be affiliate links (rel="sponsored"); we may earn a commission at no cost to you.

Get the Local AI weekly

New models worth running, what fits which GPU, and setup fixes. One short email a week, free. Unsubscribe any time.

Lila Forge Kits

Local-LLM App Starter

A tested starter for building your own app on Ollama: chat, RAG and a GPU-fit checker. Skip the setup hours.

kits.lilaforge.com → VibeEngine

Best local LLM tools, ranked

Ollama, LM Studio, Open WebUI and llama.cpp scored from public evidence, with top picks.

vibeengine.lilaforge.com →