Run gpt-oss-120b on a Dedicated Mac

Macs that run gpt-oss-120b

At 4-bit (as released) with a 32K-token context. Speeds are estimates; see real measurements on the calculator page.

• Mac Studio M4 Max 128GB Best value
  59–79 tokens/sec
  from $499/mo
  Order
• Mac Studio M5 Max 128GB
  65–86 tokens/sec
  from $569/mo Available Nov 10, 2026
  Order
• Mac Studio M5 Ultra 256GB
  109–137 tokens/sec
  from $1,199/mo Available Nov 16, 2026
  Order
• 2× Mac Studio M5 Ultra 256GB Cluster Cluster
  89–108 tokens/sec
  from $2,399/mo Available Nov 16, 2026
  Order
• Mac Studio M5 Ultra 512GB
  109–137 tokens/sec
  from $2,499/mo Available Dec 30, 2026
  Order
• 2× Mac Studio M5 Ultra 512GB Cluster Cluster
  89–108 tokens/sec
  from $4,999/mo Available Dec 30, 2026
  Order
• 4× Mac Studio M5 Ultra 512GB Cluster Cluster
  66–75 tokens/sec
  from $9,999/mo Available Dec 30, 2026
  Order

Why it runs well on a Mac

Language models have to sit entirely in memory, and generating each word means reading through them again. Apple Silicon puts up to 512 GB of fast unified memory on one chip that the GPU can use directly, so a model like this fits on one quiet machine instead of a stack of graphics cards.

gpt-oss-120b is a mixture-of-experts model: of its 117 billion parameters, only about 5.1 billion are used for each word. It needs memory for all of them but runs at the speed of a much smaller model.