Bonus: Top up now and we'll double your first deposit — get x2 credits instantly.

LLMs You Can Run on a Mac

Open-weight models with a published MLX or GGUF build that fits in 64GB of unified memory or less. Each model page shows the exact size per quantisation and the command to run it.

19 available on LLM API · 54 listed

Available Models
All Models
Model
Released
Input / 1M
Output / 1M
Context