Stable Diffusion Very Slow on Mac (M1, M2, M3)

How to fix it

Try these in order — the most common fix is first.

1. 1. Reduce the workload

   Copy steps
   1. Use an SD 1.5 model instead of SDXL
   2. Generate at 512x512 and upscale afterwards
   3. Close memory-heavy apps
   Did this fix it? Yes No

2. 2. MPS settings

   Copy steps
   1. Set PYTORCH_ENABLE_MPS_FALLBACK=1 before launching
   2. Keep your macOS and PyTorch up to date
   Did this fix it? Yes No

3. 3. Use a Mac-native tool

   Copy steps
   1. Try Draw Things or an MLX-based app, which are reported as faster on Apple Silicon
   2. Or rent a cloud GPU for heavy work
   Did this fix it? Yes No

Symptoms

A 512x512 image takes many seconds to minutes, SDXL at 1024x1024 takes 25-40 seconds or more on base chips, or the app falls back to CPU.

Why it happens

• No CUDA: PyTorch uses the slower MPS backend
• SDXL and high resolutions on base-memory Macs
• Unsupported MPS operations falling back to CPU
• Running with --no-half, which uses more memory

This guide also covers

• stable diffusion slow on mac
• stable diffusion m1 slow
• automatic1111 mac slow

The same problem in other apps

All slow, laggy or resource-hungry guides →

• Why Is Midjourney So Slow (Midjourney)

Other Stable Diffusion errors

• Stable Diffusion CUDA Out of Memory Error
• Stable Diffusion "Torch is not able to use GPU" Error
• Stable Diffusion Black Images and NaN Error
• Stable Diffusion Model Failed to Load, Exiting
• Stable Diffusion Not Using AMD GPU (Slow, CPU Only)
• Stable Diffusion xformers Error After Update
• Stable Diffusion "Couldn't install torch" Error
• Stability AI API 402 Insufficient Credits Error