| Model | Speed on your device | Memory on your device | Best fit |
|---|---|---|---|
| qwen2.5-coder:3b | Measure locally | Measure locally | Coding |
| gemma3:1b | Measure locally | Measure locally | Lightweight tasks |
| llama3.2:3b | Measure locally | Measure locally | General use |
MODEL COMPARISON
Compare trade-offs before deployment.
A good local model is one that fits your constraints: fast enough for your tasks, light on memory, and strong enough to be useful.