Model Quant Measured Estimated RAM used App Source Date
Whisper Tiny Whisper Tiny FP16 200 tok/s 200 tok/s +0% 0.1 GB Whisper Transcription editorial 2026-03-05
Qwen 2.5 0.5B Qwen 2.5 0.5B Q4 98 tok/s 40 tok/s +145% 0.6 GB LM Studio community 2026-03-05
Gemma 3 1B Gemma 3 1B Q4 62 tok/s 22 tok/s +182% 0.9 GB LM Studio community 2026-03-06