What AI models run on the Mac M1 16GB?
The Mac M1 16GB has 16 GB of unified memory (68 GB/s). About 10.7 GB is usable for a model. 168 of 298 open models we track fit at 8K context.
Apple Silicon uses unified memory; see the Mac guide for default macOS limits across machines. Other 16 GB hardware: RTX 4060 Ti 16GB, RTX 4070 Ti Super, RTX 4080, RTX 4080 Super, RTX 5060 Ti 16GB, RTX 5070 Ti, RTX 5080, RX 7800 XT.
Using a specific tool? Ollama models that run on the Mac M1 16GB · LM Studio
Models that fit, largest first
Just too big at 4-bit
These need up to 35% more memory than the Mac M1 16GB has at Q4_K_M and 8K context. A smaller quant or shorter context may work.
| Model | Q4_K_M needs | Short by |
|---|---|---|
| Moonlight 16B A3B Instruct | 10.8 GB | 0.0 GB |
| Kimi VL A3B Thinking 2506 | 10.8 GB | 0.0 GB |
| Kimi VL A3B Instruct | 10.8 GB | 0.0 GB |
| gpt-oss Safeguard 20B | 11.8 GB | 1.1 GB |
| gpt-oss 20B | 11.8 GB | 1.1 GB |
| DeepSeek MoE 16B Base | 12.5 GB | 1.7 GB |
| ERNIE 4.5 21B A3B PT | 13.6 GB | 2.9 GB |
| ERNIE 4.5 21B A3B Thinking | 13.6 GB | 2.9 GB |
| LFM2.5 24B A2B | 14.3 GB | 3.6 GB |
Questions
What is the biggest model that runs on the Mac M1 16GB?
Of the models we track, the largest that fits at 8K context is Trinity Mini (26B) at Q2_K, needing about 9.8 GB.
How much of the Mac M1 16GB's memory can a model use?
About 11 GB of 16 GB. macOS lets the GPU use about 67% of unified memory by default; this can be raised with the iogpu.wired_limit_mb setting.
Spec source: Apple specifications. Model data: Hugging Face Hub API. Speed is a bandwidth ceiling (bandwidth ÷ bytes read per token), not a benchmark. Methodology · Report an error.