AI Setup
Configure local AI for your PC2 personal cloud. With Ollama (local AI), your conversations stay 100% private -- they never leave your machine.
Installing Ollama
curl -fsSL https://ollama.com/install.sh | sh
Verify installation:
ollama --version
Recommended Models
| Model | Size | RAM Needed | Best For |
|---|---|---|---|
deepseek-r1:1.5b | 1GB | 4GB | Fast responses, basic tasks |
llama3.2:3b | 2GB | 6GB | Good balance |
phi3:mini | 2GB | 6GB | Microsoft's efficient model |
mistral:7b | 4GB | 8GB | Strong general purpose |
llama3.2:8b | 5GB | 12GB | Complex reasoning |
codellama:7b | 4GB | 8GB | Code generation |
Install a Model
ollama pull deepseek-r1:1.5b
Or via PC2: Settings -> AI Setup -> Click "Install" on any model.
List Installed Models
ollama list
GPU Acceleration
GPU dramatically improves AI speed.
NVIDIA GPUs (CUDA)
Ollama automatically uses NVIDIA GPUs if CUDA is available.
nvidia-smi
Apple Silicon (M1/M2/M3)
Ollama automatically uses Metal acceleration. No configuration needed.
No GPU?
CPU-only works for smaller models (1.5b-3b). Larger models will be slow but functional.
Connecting to Remote Ollama
If running Ollama on a different machine (like a powerful server):
On the Ollama Server:
OLLAMA_HOST=0.0.0.0 ollama serve
In PC2 Settings:
- Go to Settings -> AI Setup
- Set Ollama URL to
http://server-ip:11434 - Save
External AI Providers
PC2 also supports cloud providers for when you need more power:
| Provider | Models | Get API Key |
|---|---|---|
| OpenAI | GPT-4o, GPT-3.5 | platform.openai.com |
| Anthropic | Claude Opus 4, Claude Sonnet 4 | console.anthropic.com |
| Gemini 2.0 Flash | aistudio.google.com | |
| xAI | Grok 3, Grok 2 | xAI dashboard |
Cloud providers send data to their servers. Use Ollama for maximum privacy.
Model Recommendations by Use Case
| Use Case | Recommended Model |
|---|---|
| General chat | llama3.2:3b, mistral:7b |
| Coding | codellama:7b, deepseek-coder:6.7b |
| Writing | mistral:7b, llama3.2:8b |
| Fast responses | deepseek-r1:1.5b, phi3:mini |
Troubleshooting
"Ollama not available":
curl http://localhost:11434/api/tags
ollama serve
Slow responses: Use a smaller model (deepseek-r1:1.5b), enable GPU, or check system resources with htop.
Out of memory: Use a smaller/quantized model, close other applications, or add more RAM or swap.