AI article
How to run Qwen-Image-2.1 on a 48 GB Mac without swap
Community description: Eager loading and model CPU offload both pushed a 48 GB Mac into swap in the VAE decode. Loading one model per stage peaked at 19.0 GB with no swap. Code and numbers.
Dev.to | Oct 11, 2026 | Rodion Kazennov
Automated excerpt
In bfloat16 its text encoder takes 16. 3 GB, its transformer 13. 3 GB and its VAE 1. 3 GB, so the loaded pipeline came to 31. 4 GB on a 48 GB Mac, and decoding a 1024 × 1024 image pushed it 11 GB higher. The VAE decode then took the footprint from 32. 5 to 43. 6 GB and swap grew by 8 GB, so the guard stopped the run before it wrote an image. The footprint peaked at 19. 0 GB while encoding, at 16. 2 and 18.
Selected automatically from source text; not independently written or fact-checked. Read the original for full context.