| ▲ | Casteil 4 hours ago | |
I don't know what black magic you're up to but I see more like 30-35t/s on a 16" M5 Max using 3.8:27b Q4, regardless of whether it's mlx or gguf. qwen3.5:122b-a10b is significantly faster at around 60-65. | ||
| ▲ | syntaxing 3 hours ago | parent [-] | |
With MTP? I get 25-30 TPS on a strix halo. 50+ on a M5 max should very doable. Dflash (2) will push your TG even further | ||