Hardware requirements

Neural Junkie is lightweight as an app; the hardware story is about models, concurrency, and what you expect local agents to do. The setup wizard and Settings screen use the same RAM tiers documented here.

Neural Junkie hardware guide: RAM tiers and model sizes

App vs models

ComponentTypical size
Desktop installer~10–15 MB
Bundled Ollama runtime~1–2 GB
Default models (14B + 7B)~13.5 GB disk
Typical first-run total~15–20 GB

RAM tiers

The hub exposes GET /api/system/hardware with your tier and recommended model tags.

TierRAMDeveloper primaryBest for
minimal< 8 GBllama3.2:3b / qwen2.5-coder:7bLight chat; cloud hybrid for hard tasks
light8–15 GBqwen2.5-coder:7bSafe local dev
recommended16–31 GBqwen2.5-coder:14bFull software pack + collab
heavy32 GB+14B + LoRA basesMulti-model library, CUDA training

Multi-agent ≠ multi-model

Specialists share one Ollama backend. You are not loading six separate 14B weights. You do pay sequential inference latency and hub memory for repo indexes (capped at 2000 source files).

Hybrid strategy

Limits to know