I have been using MTPLX (replaced oMLX for me) on my M5 Pro with 64GB with Qwen3.6 27B Optimized Quality (getting about 16-19 generated tps, which is exactly double what I saw before use of MTP model and this tool) and along with OpenCode have been enjoying the experience very much :)

HUGE shoutout to Youssof! Thank you.