Rare find

Forge. A faster, easier way to run and experiment with AI image generation.

github.com/lllyasviel/stable-diffusion-webui-forge

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

June 2025
  • Fix SD upscale Batch count (#2950)
  • Add support for fp8 scaled (#2946)
  • Multithreading softinpainting (#2927)
  • Add Chroma (#2925)
  • Repair ancestral sampling for FLUX (#2915)
May 2025
  • Resolve warnings of datetime library (#2874)
  • Add BF16 to GGUF (#2877)
  • Support wd_on_output for DoRA (#2856)
  • re-add line to update sampling step (#2853)
  • Add separate cuda stream for live preview VAE (#2844)
March 2025
  • update bitsandbytes version for cuda128 support (#2712)
  • some extra lora support, inc. new glora (#2715)
February 2025
  • fix zero'd cond_t5 device (#2701)
  • this is useful (spiece.model for SD3 T5 tokenizer)(#2699)
  • SD3+ (#2688)
  • Add SDXL refiner model (#2686)
  • Add native "Stealth" infotext support (#2684)
  • increased support for custom CLIPs (#2642)
  • prompts_from_file script: add option to save grid (#2673)
  • reactivated the old hashing that work for civitai (#1647)
July 2024
  • Release —previous_versions
February 2024
  • Release —latest