---
title: HSWQ-fp8-Illustrious
canonical_url: "https://www.modelscope.cn/models/ApacheOne/HSWQ-fp8-Illustrious"
md_url: "https://www.modelscope.cn/models/ApacheOne/HSWQ-fp8-Illustrious.md"
repository: ApacheOne/HSWQ-fp8-Illustrious
last_updated: 2026-05-07
license: agpl-3.0
model_type:
  - custom
parameters: 16.4B
tensor_type:
  - F8_E4M3
  - F16
  - I64
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 81
stars: 1
---

# HSWQ-fp8-Illustrious

> HSWQ-fp8-Illustrious - ApacheOne 在 ModelScope 开源的模型。Creator: https://civitai.com/user/Bilered

ApacheOne/HSWQ-fp8-Illustrious 是 ModelScope 魔搭社区上的 16.4B 参数机器学习模型，采用 agpl-3.0 许可。

- **Repository**: ApacheOne/HSWQ-fp8-Illustrious
- **License**: agpl-3.0
- **Parameters**: 16.4B
- **Downloads**: 81
- **Stars**: 1
- **Last updated**: 2026-05-07

Source: https://www.modelscope.cn/models/ApacheOne/HSWQ-fp8-Illustrious

---

# Model info

Creator: [https://civitai.com/user/Bilered](https://civitai.com/user/Bilered)

`Lumachrome_Illustrious_HSWQ_fp8e4m3.safetensors` [https://civitai.com/models/2528730/lumachrome-illustrious](https://civitai.com/models/2528730/lumachrome-illustrious)


<table style="width: auto; border-collapse: collapse;">
  <tr>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00001_.png" target="_blank"><img src="assets/comprare_00001_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00002_.png" target="_blank"><img src="assets/comprare_00002_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00003_.png" target="_blank"><img src="assets/comprare_00003_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00004_.png" target="_blank"><img src="assets/comprare_00004_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00005_.png" target="_blank"><img src="assets/comprare_00005_.png" height="140"></a></td>
  </tr>
  <tr>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00006_.png" target="_blank"><img src="assets/comprare_00006_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00007_.png" target="_blank"><img src="assets/comprare_00007_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00008_.png" target="_blank"><img src="assets/comprare_00008_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00009_.png" target="_blank"><img src="assets/comprare_00009_.png" height="140"></a></td>
    <td style="padding: 6px;"><a href="https://huggingface.co/ApacheOne/HSWQ-fp8-Illustrious/blob/main/assets/comprare_00010_.png" target="_blank"><img src="assets/comprare_00010_.png" height="140"></a></td>
  </tr>
</table>

# Hybrid-Sensitivity-Weighted-Quantization (HSWQ)

<p align="center">
  <img src="https://raw.githubusercontent.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization/main/icon.png" width="128">
</p>

High-fidelity FP8 quantization for diffusion models (SDXL). HSWQ uses **sensitivity** and **importance** analysis instead of naive uniform cast, and offers two modes: standard-compatible (V1) and high-performance scaled (V2).

**Technical details:** [md/HSWQ_ Hybrid Sensitivity Weighted Quantization.md](https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization/blob/main/md/HSWQ_%20Hybrid%20Sensitivity%20Weighted%20Quantization.md)

**How to quantize:** [md/HSWQ_ How to quantize SDXL.md](https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization/blob/main/md/How%20to%20quantize%20SDXL.md)

**SDXL Benchmark Test Results:** [md/SDXL Benchmark Test Results.md](https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization/blob/main/test/benchmark_test.md)





# Credit & Special Acknowledgement

[https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization](https://github.com/ussoewwin/Hybrid-Sensitivity-Weighted-Quantization)

[https://github.com/tritant/ComfyUI_Kitchen_nvfp4_Converter](https://github.com/tritant/ComfyUI_Kitchen_nvfp4_Converter)

[https://github.com/NVIDIA/Model-Optimizer](https://github.com/NVIDIA/Model-Optimizer)

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](https://huggingface.co/nunchaku-tech/nunchaku-sdxl)