Comfyui / models /vae /sdxl_vae.md
v33ts's picture
Upload models/vae/sdxl_vae.md with huggingface_hub
4e779b3 verified
metadata
author: nucleardiffusion
baseModel: SDXL 1.0
hashes:
  AutoV1: 584F63AB
  AutoV2: 235745AF8D
  AutoV3: 55F20A1016E7
  BLAKE3: D46EB6988A26BCAF9E9CFA5E5C6264C4EE1A70F2018F33B8BC2DD7CA0681B490
  CRC32: 5BB75FB4
  SHA256: 235745AF8D86BF4A4C1B5B4F529868B37019A10F7C0B2E79AD0ABCA3A22BC6E1
metadata:
  format: SafeTensor
modelPage: https://civitai.com/models/140686?modelVersionId=155933
preview:
  - >-
    https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4596b7dc-4fd4-494c-bf8c-36d6a0ee89bc/width=450/2369135.jpeg
website: Civitai

Trigger Words

No trigger words

About this version

No description about this version

FIX FP16 Errors SDXL - Lower Memory use! --- sdxl-vae-fp16-fix by madebyollin

"As good as SDXL VAE but runs twice as fast and uses significantly less memory." https://huggingface.co/madebyollin/sdxl-vae-fp16-fix/discussions/7

"Same license on stable-diffusion-xl-base-1.0

same vae license on sdxl-vae-fp16-fix

Troubleshoot:

Do not use the refiner with VAE built in
Try launch param: --medvram --opt-split-attention --xformers

SDXL-VAE-FP16-Fix is the [SDXL VAE](https://huggingface.co/stabilityai/sdxl-vae, but modified to run in fp16 precision without generating NaNs.

Details:

SDXL-VAE generates NaNs in fp16 because the internal activation values are too big:

SDXL-VAE-FP16-Fix was created by finetuning the SDXL-VAE to:

  1. keep the final output the same, but
  2. make the internal activation values smaller, by
  3. scaling down weights and biases within the network

There are slight discrepancies between the output of SDXL-VAE-FP16-Fix and SDXL-VAE, but the decoded images should be close enough for most purposes." - bdsqlsz


NOT MY WORK - REUPLOADED HERE FOR EASE OF USE


COMMISSIONS NOW ACCEPTED!

I have been away saving to upgrade my pc, please help fund more work:

https://www.patreon.com/nucleardiffusion

https://ko-fi.com/nucleardiffusion