Written by Jakub Rusinowski · Last updated July 12, 2026
The original Apple Silicon chip. Up to 16 GB unified memory at 68 GB/s limits it to smaller 7–8B models in Q4. The fanless MacBook Air still works well for lightweight local AI.
| VRAM | 16 GB unified memory |
| Memory Bandwidth | 68 GB/s |
| TDP | 20 W |
| Architecture | ARM, 5nm TSMC |
| Release Year | 2020 |
| MSRP at Launch | $999 |
| Inference Speed (Llama 3.1 8B Q4_K_M) | 4.4–9.2 tok/s (estimated) |
| Inference Speed (Llama 3.3 70B Q4_K_M) | Does not fit — needs ~44 GB of 12 GB usable |
All models below run comfortably in 16 GB unified memory with Q4_K_M quantization.
| Llama 3.2 Family | Llama 3.2 11B Vision Instruct · 7 GB VRAM · Q4_K_M · llama-3-2 |
| Llama 3.1 Family | Llama 3.1 8B Instruct · 6 GB VRAM · Q4_K_M · ollama run llama3.1 |
| Qwen 2.5 Family | Qwen 2.5 14B Instruct · 9 GB VRAM · Q4_K_M · ollama run qwen2.5:14b |
| Gemma 2 Family | Gemma 2 9B IT · 6 GB VRAM · Q4_K_M · ollama run gemma2 |
| Phi-4 Mini | Phi-4 Mini (3.8B) · 3 GB VRAM · Q4_K_M · ollama run phi4-mini |
| SmolLM2 | SmolLM2 1.7B Instruct · 2 GB VRAM · Q4_K_M · ollama run smollm2:1.7b |
Install Ollama then run the recommended model for this GPU:
ollama run llama3.2:3b
Yes — the Apple M1 has 16 GB unified memory and runs The original Apple Silicon chip. Up to 16 GB unified memory at 68 GB/s limits it to smaller 7–8B models in Q4. The fanle
The Apple M1 is estimated to run Llama 3.1 8B at 4.4–9.2 tok/s with Q4_K_M quantization. Llama 3.3 70B does not fit: it needs about 44 GB against 12 GB usable. These are modelled estimates, not measurements — see /en/methodology.
With 16 GB you can run: Llama 3.2 Family, Llama 3.1 Family, Qwen 2.5 Family, Gemma 2 Family, Phi-4 Mini. Use Ollama for the easiest setup: ollama run llama3.2:3b.
← All GPU Reviews | Check Your Hardware | Full Benchmarks | Can I Run It?