Hybrid-Sensitivity-Weighted-Quantization (HSWQ)
High-fidelity ConvRot INT8 quantization for diffusion models (SDXL). HSWQ uses sensitivity and importance analysis instead of naive uniform cast. This is highly useful for users who need to strictly manage their VRAM resources while maintaining maximum image quality.
ComfyUI-compatible int8_tensorwise pack with FULL ConvRot on remaining Linear/Conv2d after DualMonitor + V4 weighted-histogram FP16 protection under a fixed 300 MiB budget. Keep ratio is 0 (r0); critical layers stay FP16 via automatic analysis, not a keep-ratio percentage. Pack path matches native_convert_int8_convrot.py.
Technical details: https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization
How to quantize (SDXL ConvRot INT8): md/HSWQ_ How to quantize SDXL.md
ComfyUI Loader for ConvRot INT8 / INT8: To load these INT8 models in ComfyUI, please use the unofficial loader node: ComfyUI-nunchaku-unofficial-loader
SDXL ConvRot INT8 Benchmark Test Results: test/benchmark_sdxl_int8.md
Benchmark (Reference)
| Model | SSIM (Avg) | File size | Compatibility |
|---|---|---|---|
| Original FP16 | 1.0000 | 100% | High |
| Naive INT8 | 0.95-0.97 | 50% | High |
| HSWQ ConvRot INT8 | 0.94-0.98 | 68% (FP16 mixed) | High (ComfyUI INT8) |
📦 Available Models
| Filename | Base Model | Version | License |
|---|---|---|---|
JANKUTrainedChenkinNoobai_v777_hswq_r32_convrot_int8.safetensors |
JANKU Trained Chenkin & Noobai-Rouwei (Illustrious-XL) | v777 | Fair AI Public License 1.0-SD |
bluePencilXL_v031_hswq_r32_convrot_int8.safetensors |
blue_pencil-XL | v0.3.1 | CreativeML Open RAIL++-M |
epicrealismXL_pureFix_hswq_r32_convrot_int8.safetensors |
epiCRealism XL | pureFix | CreativeML Open RAIL++-M |
oneObsession_v23_hswq_r32_convrot_int8.safetensors |
OneObsession | v23 | CreativeML Open RAIL++-M |
waiIllustriousSDXL_v170_hswq_r32_convrot_int8.safetensors |
Illustrious-XL v1.7 (WAI-illustrious-SDXL) | v17.0 (HF weight) | Fair AI Public License 1.0-SD |
📜 Credits & License
🏆 Special Acknowledgement
We extend our deepest respect and gratitude to the Nunchaku Team for their groundbreaking work on SVDQ quantization and for sharing their models with the community. This collection relies heavily on their research and original implementation.
- Original Repository: nunchaku-tech/nunchaku-sdxl
Base Models
These models are derivatives of their respective creators. All credit for aesthetic tuning and model training belongs to the original creators.
- JANKU Trained Chenkin & Noobai-Rouwei (Illustrious-XL): Created by janxd.
- blue_pencil-XL: Created by Euge_us.
- epiCRealism XL: Created by epinikion.
- OneObsession: Created by Polyhedron.
- WAI-illustrious-SDXL: Created by WAI0731.
Disclaimer: These models are provided for optimization and research purposes. Please adhere to the original licenses of the base models.