Outcomes, not parts
Local AI Stack Recipes
Proven, working, deployable setups. Each recipe is a complete blueprint - the exact tools, models, and hardware - for a real outcome you can stand up today. Runs on your own machine, $0/mo vs cloud.
Working docker-composeReal hardware + costYour data never leaves your network
| Name | Category | Cost | Setup | Featured |
|---|---|---|---|---|
| Local AI App Builder (Dify + Ollama) | - | - | ||
| Local AI Workflow Automation (n8n + Ollama) | - | - | ||
| Local Cursor (AI coding) | - | - | ||
| Local Document RAG (AnythingLLM + Ollama) | - | - | ||
| Local Image Generation (ComfyUI + FLUX) | - | - | ||
| Mac Mini Local AI | - | - | ||
| Private ChatGPT | - | - | ||
| Qwen3.6 Local Multimodal Chat | - | - | ||
| Terminal AI Coding (Aider + Ollama) | - | - | ||
| Tool-Enabled AI Chat (Open WebUI + MCP) | - | - | ||
| Windows Desktop AI with RAG | - | - | ||
| Apple Silicon Local Agents | - | - | - | |
| DiffusionGemma Local | - | - | - | |
| Gemma 4 QAT on 10GB Laptop | - | - | - | |
| Gemma 4 QAT on 16GB Laptop | - | - | - | |
| GGUF Local Model Hub | - | - | - | |
| Local Gemini Alternative | - | - | - | |
| Local LLM Research Hub | - | - | - | |
| Local Voice Assistant (Whisper + Ollama + Kokoro TTS) | - | - | - | |
| Multi-Model API Gateway (LiteLLM + Ollama) | - | - | - | |
| Off-Grid CrankGPT | - | - | - | |
| Ollama 0.30 GPU Boost | - | - | - | |
| Ollama Mac Metal AI | - | - | - | |
| Ollama WebUI Docker | - | - | - | |
| Qwen 3.5 9B Local Apple | - | - | - | |
| Visual AI Agent Builder (Langflow + Ollama) | - | - | - |