model · Alibaba (Qwen team)

Qwen3.6 (35B-A3B)

Qwen3.6-35B-A3B is Alibaba's open mixture-of-experts model: 35B total parameters but only ~3B active per token, so it generates quickly while keeping the quality of a much larger model. It has a 256K context window, strong coding and multilingual skills, an Apache-2.0 licence, and first-class Apple MLX support — at 4-bit it needs about 19 GB, making a 32 GB Mac its natural home. It is the standout 'serious model on a Mac' pick.

  • free
  • self-host
  • Commercial

What it can help with

  • mixture-of-experts
  • apple-silicon
  • mlx-support
  • 256k-context
  • code-generation
  • multilingual
  • 4-bit-quantization
  • apache-2.0