Will it run? / Code Llama 34B / Mac mini M6 (32GB)

Can the Mac mini M6 (32GB) run Code Llama 34B?

Usable, but slow Yes, at about 7.2 tokens/sec (estimated). It works, but you'll wait on longer answers.

7.2estimated tokens/sec
20.5 GBmemory needed
21.4 GBavailable to the model
170 GB/smemory bandwidth
ollama run codellama:34b

Check the Mac mini M6 (32GB) price →

How we got this number

Each generated token reads the whole model from memory once: 19 GB. The Mac mini M6 (32GB) moves 170 GB/s, so the ceiling is 8.9 tokens/sec. Real machines reach a fraction of that ceiling; we use 80%, calibrated against a measured Mac run. Send us your real numbers and we'll replace the estimate.

Faster machines for Code Llama 34B

Mac Studio M5 Ultra (96GB)52 tok/s$5,499Check price →
Mac Studio M5 Ultra (256GB)52 tok/s$9,499Check price →
Mac Studio M5 Max (40-core GPU, 64GB)26 tok/s$3,799Check price →
Mac mini M5 Pro (64GB)13 tok/sCheck price →

Other models on the Mac mini M6 (32GB)

lfm 2.5 8B (A1B)Fast111 tok/s
Qwen 3 30B (A3B)Fast40 tok/s
Qwen3 Coder 30B (A3B)Fast40 tok/s
Llama 2 7BFast36 tok/s
Code Llama 7BFast36 tok/s
DeepSeek Coder 6.7BFast36 tok/s