From 01567526a2b55842c29c04e398d8c200fe095d07 Mon Sep 17 00:00:00 2001 From: Lmxyy1999 Date: Sun, 16 Nov 2025 02:24:27 +0000 Subject: [PATCH] Upload ./README.md to ModelScope hub --- README.md | 68 +++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 54 insertions(+), 14 deletions(-) diff --git a/README.md b/README.md index aca65b4..09c5b44 100644 --- a/README.md +++ b/README.md @@ -37,7 +37,6 @@ This repository contains Nunchaku-quantized versions of [Qwen-Image](https://hug - [2025-08-27] 🔥 Release **4-bit [4/8-step lightning Qwen-Image](https://huggingface.co/lightx2v/Qwen-Image-Lightning)**! - [2025-08-15] 🚀 Release 4-bit SVDQuant quantized Qwen-Image model with rank 32 and 128! - ## Model Details ### Model Description @@ -49,18 +48,59 @@ This repository contains Nunchaku-quantized versions of [Qwen-Image](https://hug ### Model Files -- [`svdq-int4_r32-qwen-image.safetensors`](./svdq-int4_r32-qwen-image.safetensors): SVDQuant INT4 (rank 32) Qwen-Image model. For users with non-Blackwell GPUs (pre-50-series). -- [`svdq-int4_r128-qwen-image.safetensors`](./svdq-int4_r128-qwen-image.safetensors): SVDQuant INT4 (rank 128) Qwen-Image model. For users with non-Blackwell GPUs (pre-50-series). It offers better quality than the rank 32 model, but it is slower. -- [`svdq-int4_r32-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-int4_r32-qwen-image-lightningv1.0-4steps.safetensors): SVDQuant INT4 (rank 32) 4-step Qwen-Image model by fusing [Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) using LoRA strength = 1.0. For users with non-Blackwell GPUs (pre-50-series). -- [`svdq-int4_r128-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-int4_r128-qwen-image-lightningv1.0-4steps.safetensors): SVDQuant INT4 (rank 128) 4-step Qwen-Image model by fusing [Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) using LoRA strength = 1.0. For users with non-Blackwell GPUs (pre-50-series). -- [`svdq-int4_r32-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-int4_r32-qwen-image-lightningv1.1-8steps.safetensors): SVDQuant INT4 (rank 32) 8-step Qwen-Image model by fusing [Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) using LoRA strength = 1.0. For users with non-Blackwell GPUs (pre-50-series). -- [`svdq-int4_r128-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-int4_r128-qwen-image-lightningv1.1-8steps.safetensors): SVDQuant INT4 (rank 128) 8-step Qwen-Image model by fusing [Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) using LoRA strength = 1.0. For users with non-Blackwell GPUs (pre-50-series). -- [`svdq-fp4_r32-qwen-image.safetensors`](./svdq-fp4_r32-qwen-image.safetensors): SVDQuant NVFP4 (rank 32) Qwen-Image model. For users with Blackwell GPUs (50-series). -- [`svdq-fp4_r128-qwen-image.safetensors`](./svdq-fp4_r128-qwen-image.safetensors): SVDQuant NVFP4 (rank 128) Qwen-Image model. For users with Blackwell GPUs (50-series). It offers better quality than the rank 32 model, but it is slower. -- [`svdq-fp4_r32-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-fp4_r32-qwen-image-lightningv1.0-4steps.safetensors): SVDQuant NVFP4 (rank 32) 4-step Qwen-Image model by fusing [Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) using LoRA strength = 1.0. For users with Blackwell GPUs (50-series). -- [`svdq-fp4_r128-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-fp4_r128-qwen-image-lightningv1.0-4steps.safetensors): SVDQuant NVFP4 (rank 128) 4-step Qwen-Image model by fusing [Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) using LoRA strength = 1.0. For users with Blackwell GPUs (50-series). -- [`svdq-fp4_r32-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-fp4_r32-qwen-image-lightningv1.1-8steps.safetensors): SVDQuant NVFP4 (rank 32) 8-step Qwen-Image model by fusing [Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) using LoRA strength = 1.0. For users with Blackwell GPUs (50-series). -- [`svdq-fp4_r128-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-fp4_r128-qwen-image-lightningv1.1-8steps.safetensors): SVDQuant NVFP4 (rank 128) 8-step Qwen-Image model by fusing [Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) using LoRA strength = 1.0. For users with Blackwell GPUs (50-series). +**Data Type**: `INT4` for non-Blackwell GPUs (pre-50-series), `NVFP4` for Blackwell GPUs (50-series). +**Rank**: `r32` for faster inference, `r128` for better quality but slower inference. + + + + + +### Base Models + +Standard inference speed models for general use + + +| Data Type | Rank | Model Name | Comment | +|-----------|------|----------|---------| +| INT4 | r32 | [`svdq-int4_r32-qwen-image.safetensors`](./svdq-int4_r32-qwen-image.safetensors) | | +| | r128 | [`svdq-int4_r128-qwen-image.safetensors`](./svdq-int4_r128-qwen-image.safetensors) | | +| NVFP4 | r32 | [`svdq-fp4_r32-qwen-image.safetensors`](./svdq-fp4_r32-qwen-image.safetensors) | | +| | r128 | [`svdq-fp4_r128-qwen-image.safetensors`](./svdq-fp4_r128-qwen-image.safetensors) | | + + + + + +### 4-Step Distilled Models + +4-step distilled models fused with [Qwen-Image-Lightning-4steps-V1.0 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) using LoRA strength = 1.0 + + +| Data Type | Rank | Model Name | Comment | +|-----------|------|----------|---------| +| INT4 | r32 | [`svdq-int4_r32-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-int4_r32-qwen-image-lightningv1.0-4steps.safetensors) | Fused with [Qwen-Image-Lightning-4steps-V1.0 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) | +| | r128 | [`svdq-int4_r128-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-int4_r128-qwen-image-lightningv1.0-4steps.safetensors) | Fused with [Qwen-Image-Lightning-4steps-V1.0 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors). Better quality, slower inference | +| NVFP4 | r32 | [`svdq-fp4_r32-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-fp4_r32-qwen-image-lightningv1.0-4steps.safetensors) | Fused with [Qwen-Image-Lightning-4steps-V1.0 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors) | +| | r128 | [`svdq-fp4_r128-qwen-image-lightningv1.0-4steps.safetensors`](./svdq-fp4_r128-qwen-image-lightningv1.0-4steps.safetensors) | Fused with [Qwen-Image-Lightning-4steps-V1.0 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-4steps-V1.0-bf16.safetensors). Better quality, slower inference | + + + + + +### 8-Step Distilled Models + +8-step distilled models fused with [Qwen-Image-Lightning-8steps-V1.1 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) using LoRA strength = 1.0 + + +| Data Type | Rank | Model Name | Comment | +|-----------|------|----------|---------| +| INT4 | r32 | [`svdq-int4_r32-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-int4_r32-qwen-image-lightningv1.1-8steps.safetensors) | Fused with [Qwen-Image-Lightning-8steps-V1.1 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) | +| | r128 | [`svdq-int4_r128-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-int4_r128-qwen-image-lightningv1.1-8steps.safetensors) | Fused with [Qwen-Image-Lightning-8steps-V1.1 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors). Better quality, slower inference | +| NVFP4 | r32 | [`svdq-fp4_r32-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-fp4_r32-qwen-image-lightningv1.1-8steps.safetensors) | Fused with [Qwen-Image-Lightning-8steps-V1.1 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors) | +| | r128 | [`svdq-fp4_r128-qwen-image-lightningv1.1-8steps.safetensors`](./svdq-fp4_r128-qwen-image-lightningv1.1-8steps.safetensors) | Fused with [Qwen-Image-Lightning-8steps-V1.1 LoRA](https://huggingface.co/lightx2v/Qwen-Image-Lightning/blob/main/Qwen-Image-Lightning-8steps-V1.1-bf16.safetensors). Better quality, slower inference | + + + ### Model Sources @@ -68,7 +108,7 @@ This repository contains Nunchaku-quantized versions of [Qwen-Image](https://hug - **Inference Engine:** [nunchaku](https://github.com/nunchaku-tech/nunchaku) - **Quantization Library:** [deepcompressor](https://github.com/nunchaku-tech/deepcompressor) - **Paper:** [SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models](http://arxiv.org/abs/2411.05007) -- **Demo:** [svdquant.mit.edu](https://svdquant.mit.edu) +- **Demo:** [demo.nunchaku.tech](https://demo.nunchaku.tech) ## Usage