MLX

Optimisations Wednesday, 25% more throughput than Ollama↗

Smarter & deeper MTP, MTPLX, and less CPU/GPU synchronisation on Qwen.

• MLX
• Qwen
• Performance