Automatic1111 for Apple metal, 40% speed up sd1.5

vanillax 1 hour ago |

SD 1.5 is so insanely old at this point im not sure the point of this when you have Mini Max H3 on metal https://news.ycombinator.com/item?id=49252179

dmikeyanderson 24 minutes ago | |

old but relevant, it's like a production system. We can continue to tune and squeeze 1mp pictures out of it in high quality on relevant localized hardware without breaking a sweat. 1-3 seconds on a m3 system, and then cascade of tooling below stays the same.

SV_BubbleTime 57 minutes ago | |

A1111 also had its time. Comfy takes a couple hours to learn, and then it’s vastly superior. Like, if you are already on Mac there is no reason to also kick yourself in the face trying to go back to 1111.

qclibre22 42 minutes ago | | |

There's also sdcpp ( https://github.com/leejet/stable-diffusion.cpp ) . It is for people that know command line shells like bash or Powershell. SDCPP is like llama.cpp but it is for using text to image AI models. Not everyone likes the node/graph programming interface in Comfy or a Webpage interface. Some people need these GUI/web interfaces and get lost and frustrated in shell scripting.

dmikeyanderson 26 minutes ago | | |

Totally and sdcpp and drawthings were inspiration for this work. It's not just about the UI, there's a whole stack of plugins, APIs etc. It would be a bigger shift, than just use other UI. Composition etc.

Plus this was fun to learn where the attention and time spent is eating up. I plan to git into the UNet, and then do some Multi-Model Memory Attention tuning as well, for Refining and Inpaint Swaps for the 8Gb Macs.

jurgenburgen 1 hour ago |

Trigger warning: AI content.

trencedamp 47 minutes ago | |

If anyone is triggered by AI content they surely avoid HN

IronWolve 1 hour ago |

Automatic1111 wasnt being updated, people moved to Forge Neo, ComfyUI and Maestro.

vunderba 2 minutes ago | |

Invoke AI is also really good if you prefer working in a canvas-style environment, in a more traditional graphical way. It easily has the nicest workflows for outpainting/regional/inpainting by far.

https://github.com/invoke-ai/invokeai

dmikeyanderson 29 minutes ago | |

Yup! It was a fun experiment though!

Der_Einzige 26 minutes ago | |

What a crazy repo. Some cyber archeologists need to do a case analysis of the entire diffusion model ecosystem circa the original Stable Diffusion up to now.

smallerize 1 hour ago |

The speedup includes running LoRAs, right? Which ones are you using?

dmikeyanderson 29 minutes ago | |

Yes it includes LoRA composition as well, a bunch from the internet, and they work just fine!