Model Quant Measured Estimated RAM used App Source Date
Qwen 2.5 0.5B Qwen 2.5 0.5B Q8 185 tok/s 100 tok/s +85% 0.9 GB Ollama editorial 2026-02-20
Gemma 3 1B Gemma 3 1B Q8 96 tok/s 55 tok/s +75% 1.5 GB LM Studio community 2026-02-22
Whisper Large V3 Whisper Large V3 Q8 78 tok/s 39 tok/s +100% 1.8 GB MacWhisper editorial 2026-03-12
Llama 3.2 3B Llama 3.2 3B Q4 65 tok/s 18 tok/s +261% 2.1 GB Jan community 2026-03-10
Phi-4 Mini Phi-4 Mini Q4 58 tok/s 15 tok/s +287% 2.4 GB LM Studio editorial 2026-02-20