Explore AI
Mac with 64 GB or more
You can run almost anything that is open, at some speed.
You can run almost anything that is open, at some speed.
- Platform
- Apple Silicon
- Memory
- 64 GB unified
Large models fit, but generation speed drops as models grow. Token rate is the wall here, not capacity.
Runs comfortably
- Qwen3 32B 24 GB at Q4 Apache-2.0
- Gemma 4 31B 24 GB at Q4 Apache-2.0
- GPT-OSS 120B 64 GB at Q4 · 120B total, 5.1B active Apache-2.0
- Qwen3 Coder 30B A3B 20 GB at Q4 · 30B total, 3B active Apache-2.0
Worth knowing
Memory bandwidth is the real limiter on Apple Silicon, not capacity. A 120B model may fit and still feel slow.