CyberRealistic XL v9.0 β€” GGUF

This repository contains the extracted components and GGUF quantizations of CyberRealistic XL v9.0 by Cyberdelia, intended for use with ComfyUI-GGUF.

The scripts and workflows used to generate these files are documented in https://github.com/insecure-erasure/safetensors-goodies.

SDXL Pipeline Comparison: Checkpoint vs GGUF + VAE Configurations

This section validates that different SDXL pipeline configurations produce equivalent results.

Full checkpoint F16 GGUF F16 + VAE from checkpoint GGUF F16 + VAE fp16-fix F16
Full checkpoint β€” F16
Standard SDXL checkpoint with the VAE baked in at F16. Reference output.
GGUF F16 + VAE extracted from checkpoint
UNet in GGUF F16, VAE and CLIP extracted from the same checkpoint in F16. Pixel-perfect identical to the reference.
GGUF F16 + madebyollin VAE fp16-fix β€” F16
Same as above but with madebyollin's VAE (originally FP32, converted to F16).

The following table shows the difference between some of the K-quants available with llama-quantize from llama.cpp using city96's diffusion model patch.

Q8_0 Q6_K Q4_K_M
Q8_0 Q6_K Q4_K_M

All images were generated using the same prompt, seed, and sampler settings.

  • Checkpoint: cyberrealisticXL v9.0 by Cyberdelia
  • Prompt: ultra detailed bust portrait of a 28-year-old woman, medium shot, wearing turtle neck sweater, subject centered, natural framing, 50mm lens, realistic skin texture, detailed eyes, soft cinematic lighting, professional photography, high dynamic range, sharp focus, natural colors
  • Negative prompt: lowres, blurry, oversharpened, jpeg artifacts, bad anatomy, extra fingers, deformed hands, cross-eye, plastic skin, overexposed, underexposed, watermark, text, logo
  • Sampler: dpmpp_2m
  • Scheduler: karras
  • Steps: 30
  • Denoise: 1.0

Repository Structure

/
β”œβ”€β”€ cyberrealisticXL_v90-F16.gguf
β”œβ”€β”€ cyberrealisticXL_v90-Q8_0.gguf
β”œβ”€β”€ cyberrealisticXL_v90-Q6_K.gguf
β”œβ”€β”€ cyberrealisticXL_v90-Q5_K_S.gguf
β”œβ”€β”€ cyberrealisticXL_v90-Q4_K_M.gguf
β”œβ”€β”€ cyberrealisticXL_v90-Q3_K_M.gguf
β”œβ”€β”€ cyberrealisticXL_v90-Q2_K.gguf
└── safetensors/
    β”œβ”€β”€ cyberrealisticXL_v90_unet.safetensors
    β”œβ”€β”€ cyberrealisticXL_v90_clip_l.safetensors
    β”œβ”€β”€ cyberrealisticXL_v90_clip_g.safetensors
    └── cyberrealisticXL_v90_vae.safetensors

Credits

This repository contains derived weights from CyberRealistic XL v9.0 by Cyberdelia, used under its original license.

Downloads last month
85
GGUF
Model size
3B params
Architecture
sdxl
Hardware compatibility
Log In to add your hardware

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for InsecureErasure/CyberRealisticXL-GGUF

Quantized
(1)
this model