Mac 18 GB for local AI
Which image and video models run on an Apple Silicon Mac with 18 GB of unified memory, which file to download for each, and how much of the memory they need.
Specs: support.apple.com
01What runs on it
| Model | Verdict | Best file | Size | Needed |
|---|---|---|---|---|
| image models | ||||
| FLUX.1 [dev] | Runs | Q6_K | 9.9 GB | 12.2 GB |
| FLUX.1 [schnell] | Runs | Q6_K | 9.8 GB | 12.1 GB |
| FLUX.1 Kontext [dev] | Runs | Q6_K | 9.8 GB | 12.4 GB |
| FLUX.1 Krea [dev] | Runs | Q6_K | 9.8 GB | 12.1 GB |
| FLUX.1 Fill [dev] | Runs | Q6_K | 9.9 GB | 12.5 GB |
| FLUX.2 [dev] | Offload only | Q2_K | 12.9 GB | 16.2 GB |
| FLUX.2 [klein] 9B | Runs well | Q8_0 | 10.0 GB | 12.3 GB |
| FLUX.2 [klein] 4B | Runs well | 16-bit | 7.8 GB | 9.6 GB |
| Krea 2 (Turbo) | Runs | Q6_K | 10.6 GB | 13.2 GB |
| Qwen-Image | Tight | Q3_K_M | 9.7 GB | 12.5 GB |
| Qwen-Image-Edit (2511) | Tight | Q3_K_M | 9.9 GB | 12.9 GB |
| Qwen-Image 2.1 | Runs well | Q8_0 | 7.6 GB | 9.9 GB |
| Z-Image Turbo | Runs well | 16-bit | 12.3 GB | 14.3 GB |
| Z-Image (base) | Runs well | 16-bit | 12.3 GB | 14.3 GB |
| Ideogram 4 | Offload only | Q4_1 | 6.2 GB | 14.7 GB |
| Boogu-Image (Turbo) | Runs well | Q8_0 | 11.6 GB | 14.2 GB |
| ERNIE-Image (Turbo) | Runs well | Q8_0 | 8.7 GB | 11.0 GB |
| HiDream-O1-Image | Offload only | 16-bit | 16.4 GB | 19.2 GB |
| Mage-Flow (Microsoft) | Runs well | 16-bit | 8.2 GB | 10.0 GB |
| Lumina Image 2.0 | Runs well | 16-bit | 5.2 GB | 7.0 GB |
| HiDream-I1 (Full) | Runs | Q4_K_M | 11.5 GB | 14.3 GB |
| HiDream-I1 (Dev) | Runs | Q4_K_M | 11.5 GB | 14.3 GB |
| Stable Diffusion 3.5 Large | Runs well | Q8_0 | 8.8 GB | 11.1 GB |
| Stable Diffusion 3.5 Medium | Runs well | 16-bit | 5.1 GB | 6.9 GB |
| Chroma1-HD | Runs well | Q8_0 | 9.7 GB | 12.0 GB |
| SDXL 1.0 | Runs well | 16-bit | 6.9 GB | 7.1 GB |
| Illustrious XL / Pony (SDXL anime) | Runs well | 16-bit | 6.9 GB | 7.1 GB |
| Stable Diffusion 1.5 | Runs well | 16-bit | 2.1 GB | 3.3 GB |
| HunyuanImage 2.1 | Tight | Q3_K_M | 9.8 GB | 13.1 GB |
| video models | ||||
| Wan 2.1 T2V 14B | Tight | Q3_K_M | 7.6 GB | 11.9 GB |
| Wan 2.1 T2V 1.3B | Runs well | 16-bit | 2.8 GB | 5.6 GB |
| Wan 2.1 I2V 14B 480P | Tight | Q3_K_M | 8.6 GB | 12.9 GB |
| Wan 2.1 I2V 14B 720P | Offload only | Q3_K_M | 8.6 GB | 15.4 GB |
| Wan 2.1 VACE 14B | Tight | Q3_K_S | 7.8 GB | 12.6 GB |
| Wan 2.2 T2V A14B | Runs | Q4_K_M | 9.7 GB | 14.0 GB |
| Wan 2.2 I2V A14B | Runs | Q4_K_M | 9.7 GB | 14.0 GB |
| Wan 2.2 TI2V 5B | Runs well | 16-bit | 10.0 GB | 13.8 GB |
| Wan 2.2 Animate 14B | Tight | Q3_K_M | 8.6 GB | 13.9 GB |
| Wan Animate 2 (14B) | Tight | Q3_K_M | 8.6 GB | 13.9 GB |
| Wan 2.2 S2V 14B | Tight | Q2_K | 9.5 GB | 14.3 GB |
| SCAIL-2 (character animation) | Tight | Q2_K | 7.3 GB | 12.6 GB |
| HunyuanVideo (13B, original) | Runs | Q5_K_M | 9.4 GB | 13.7 GB |
| LTX-Video 13B (0.9.8) | Runs | Q5_K_M | 9.8 GB | 14.1 GB |
| HunyuanVideo 1.5 | Runs well | Q8_0 | 9.0 GB | 13.3 GB |
| LTX-2 (19B) | Tight | Q2_K | 8.1 GB | 12.9 GB |
| LTX-2.3 (22B) | Tight | Q2_K | 8.3 GB | 13.1 GB |
| LTX-2.5 (22B) | Tight | Q2_K | 8.8 GB | 13.6 GB |
| MiniMax H3 (33B) | Offload only | Q4_K_M | 19.9 GB | 25.7 GB |
| MiniMax H3 Pruned | Offload only | Q3_K_M | 8.9 GB | 14.7 GB |
Calculated from real file sizes plus working memory. How the numbers work.
02Good to know
A Mac with 18 GB shares that memory between the CPU and the GPU. On current macOS the GPU may use about 14.4 GB of it by default (older macOS versions: about 12.9 GB); that is the figure used here. ComfyUI can go past it, but macOS then starts compressing and swapping memory and everything slows down. ComfyUI runs on Apple GPUs through PyTorch's MPS backend. 16-bit and GGUF files work; FP8 and INT8 files do not save memory on a Mac, so they are skipped. Speed is the catch: even the fastest Macs are several times slower per image than a desktop RTX card. Chips sold with 18 GB: M3 Pro (memory bandwidth 150–150 GB/s — the higher, the faster). Everything about Macs and local AI →
03Measured and reported results
No measured results yet. Send yours.