Model Quant Measured Estimated RAM used App Source Date
Qwen 2.5 0.5B Qwen 2.5 0.5B Q8 145 tok/s 113 tok/s +28% 0.6 GB Ollama editorial 2026-03-20
Phi-4 Mini Phi-4 Mini Q4 52 tok/s 17 tok/s +206% 2.4 GB LM Studio editorial 2026-03-20