Run AI on your own computer
Private, free to run, and works offline. Find the model that fits your GPU, follow tested setup guides, and buy only the hardware you need.
Start here
GPU model picker
VRAM + RAM + what you want to do → Ollama models that fit, with estimated speeds.
BasicsVRAM & quantisation explained
Why a 9B model fits in 8GB, and what Q4_K_M actually means.
HardwareGPU buying guide by tier
8GB, 12GB, 16GB, 24GB and mini PCs: what each can run.
Guides
How to install Ollama on Windows (2026)
Install Ollama on Windows 10 or 11, run your first local model, check it's using your GPU, and move models to another drive.
How to install Ollama on Linux (Ubuntu, Debian, Fedora, Arch)
One-line install, NVIDIA and AMD GPU setup, running Ollama as a systemd service, and exposing it on your network safely.
How to install Ollama on a Mac (Apple Silicon)
Install Ollama on macOS, pick models that fit your unified memory, and know where files live.
LM Studio vs Ollama: which should you use? (2026)
An honest comparison of the two most popular local AI apps: interface, licence, Mac MLX support, headless servers and integrations.
VRAM and quantisation explained: what actually fits on your GPU
Why a 9B model fits in 8GB, what Q4_K_M means, how context length eats memory, and why bandwidth sets your speed.
Best local AI models for an 8GB GPU (2026)
The Ollama models that fit fully on an 8GB card like the RTX 4060, RTX 5060 or RX 7600, by use case, with exact tags.
Best local AI models for a 12GB GPU (2026)
What runs on an RTX 3060 12GB, RTX 5070 or Arc B580: the best 9B-14B Ollama models by use case.
Best local AI models for a 16GB GPU (2026)
16GB cards like the RTX 5060 Ti 16GB and RX 9070 XT unlock 20-24B models: gpt-oss 20B, Mistral Small 3.2, Devstral Small 2.
Best local AI models for a 24GB GPU (2026)
RTX 3090, 4090 or RX 7900 XTX? Here are the best 27-35B Ollama models that run fully on 24GB, from Qwen3.6 to Gemma 4 31B.
Set up a free local coding assistant (Continue, VS Code, Cursor)
Use Ollama models for chat, edits and tab autocomplete in VS Code with Continue, plus the honest truth about local models in Cursor.
Chat with your documents locally: a simple RAG setup with Ollama
Build private document Q&A with Ollama: no-code options (Open WebUI, LM Studio) and a 40-line Python version you can understand.
Running local AI on AMD, Intel Arc and mini PCs
Which AMD Radeon and Intel Arc GPUs work with Ollama, ROCm vs Vulkan, and how unified-memory mini PCs like the Ryzen AI Max+ 395 compare.
Why run AI locally? Privacy, cost and control explained
What local AI actually protects, where it doesn't, when the cost works out, and how to keep a local setup genuinely private.
Hardware by GPU tier
8GB GPUs
The entry tier. Runs 4B-9B models fully on the GPU at Q4 quantisation, which covers everyday chat, summarising and light coding.
12GB GPUs
The sweet spot for 8B-14B models with room for longer context. 12B models like Gemma 4 12B fit fully on the GPU.
16GB GPUs
Unlocks 20-24B models such as gpt-oss 20B, Mistral Small 3.2 and Devstral Small 2, plus 14B models with long context.
24GB+ GPUs
Runs the best local models: Qwen3.6 27B, Gemma 4 31B and the 30-35B mixture-of-experts models, fully on the GPU.
Mini PCs & unified memory
Machines where the GPU shares system memory. Slower per token than a big GPU, but they can hold far larger models in a small, quiet box.
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.
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 → VibeEngineBest local LLM tools, ranked
Ollama, LM Studio, Open WebUI and llama.cpp scored from public evidence, with top picks.
vibeengine.lilaforge.com →