SDXL Base 1.0 ConvRot (INT8, INT6 & INT4) for ComfyUI

ConvRot-quantized single-file ComfyUI checkpoints of stabilityai/stable-diffusion-xl-base-1.0 in three clean bit widths: Full INT8, Full INT6 (w6a8_int8), and Full INT4 (asym_w4a8_int8).

Built with ConvRot (block Hadamard rotation) + 4-stage sequential Act-Order GPTQ (H_qq) + Hessian-weighted per-row clip search (INT8) / through-grid FP8 group-scale search, damped coordinate-descent code refinement & Lloyd-Max codebooks + cumulative activation-drift bias correction, keeping Conv2d layers, text encoders (CLIP-L / OpenCLIP-G), and VAE unquantized in FP16.

Related SDXL ConvRot repositories:


Available Checkpoints

Precision Tier File (ComfyUI/models/checkpoints/) Size (vs 6.94 GB FP16) Quantized UNet Linears
Full INT8 ConvRot sdxl-base-1.0-int8-convrot.safetensors 4.72 GB (-32.0%) 739 / 739 INT8
Full INT6 ConvRot sdxl-base-1.0-int6-convrot.safetensors 4.31 GB (-37.8%) 658 / 739 INT6 (W6A8) + 81 INT8 (K=640/det)
Full INT4 ConvRot sdxl-base-1.0-int4-convrot.safetensors 3.77 GB (-45.6%) 658 / 739 INT4 (W4A8) + 81 INT8 (K=640/det)

All files load directly in stock ComfyUI (Load Checkpoint, ComfyUI >= v0.45.0) using native comfy-kitchen CUDA kernels โ€” no extra nodes or unpacking required. (For SDXL Base/Lightning/Turbo, 80 small K=640 linear layers plus the single architecture-detection to_k key remain in ConvRot INT8 because comfy-kitchen w6a8/w4a8 kernels require K % 256 == 0.)


Visual Comparisons (50 steps, normal, 1024x1024)

Base Comparison p0 Base Comparison p2 Base Comparison p5 Base Comparison p6


Evaluation Metrics (8 prompts x 1024x1024, ComfyUI v0.45.3)

Measured via scripts/verify_sdxl_quant.py directly against unquantized FP16 (sd_xl_base_1.0.safetensors):

  • Unquantized BF16 Reference (vs FP16): Cosine 0.99186 (min 0.96847), SSIM 0.94269 (min 0.85051), PSNR 32.20 dB (min 26.31 dB)
Precision Tier Cosine Mean / Min SSIM Mean / Min PSNR Mean / Min Warm Speed/Img Report
Full INT8 ConvRot 0.99459 / 0.98156 0.95612 / 0.89355 33.28 / 27.24 dB 4.76s g4_base_int8_summary.json
Full INT6 ConvRot 0.99381 / 0.98645 0.95011 / 0.91579 32.19 / 28.54 dB 5.32s g4_base_int6_summary.json
Full INT4 ConvRot 0.97411 / 0.96053 0.87257 / 0.82174 26.18 / 22.86 dB 5.40s g4_base_int4_summary.json

How to Use in ComfyUI

  1. Download INT8, INT6, or INT4 .safetensors into ComfyUI/models/checkpoints/.
  2. Load directly with ComfyUI's built-in Load Checkpoint node (ComfyUI >= v0.45.0).

License & Use Restrictions (CreativeML Open RAIL++-M)

All checkpoints in this repository are Derivatives of the Model under the CreativeML Open RAIL++-M License (Copyright ยฉ 2023 Stability AI). See NOTICE for full attributions and modification details.

Under Section III, Paragraph 5 & Attachment A of LICENSE.md, all users and downstream recipients are legally bound by the Open RAIL++-M Use-Based Restrictions and must pass through LICENSE.md and NOTICE with any redistribution.

Downloads last month
181
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for SUSSYEATING/SDXL-Base-1.0-ConvRot

Quantized
(26)
this model