---
license: other
license_name: qwen-research
license_link: LICENSE
base_model:
  - Qwen/Qwen-Image-2.1
  - Viggle/Qwen-Image-2.1-viggle-turbo
pipeline_tag: text-to-image
language:
  - en
  - zh
tags:
  - comfyui
  - qwen-image-2.1
  - lora
  - image-editing
  - six-step
---

# Qwen-Image-2.1 Viggle Turbo · ComfyUI LoRA

**Built with Qwen.** [Viggle](https://huggingface.co/Viggle/Qwen-Image-2.1-viggle-turbo) 训练，T8star 转换。当前提供 **v0.2.1 六步 r256 / r128**；仓库名称中的 `4step-r64` 是旧版名称，旧权重保留供复现。

> **v0.2.1 必须按随附工作流使用 ComfyUI 内置 `LoraLoaderBypassModelOnly`，搜索名称：`Load LoRA (Bypass, Model Only) (for debugging)`。强度设为 `1.0`，搭配 BF16 底座；无需安装新节点。** 普通 `LoraLoader` / `LoraLoaderModelOnly` 会合并更新，存在作者指出的舍入精度损失。找不到该内置节点时，请更新 ComfyUI。

> **For the supplied v0.2.1 workflow, use ComfyUI's built-in `LoraLoaderBypassModelOnly`: search for `Load LoRA (Bypass, Model Only) (for debugging)`. Set strength to `1.0` and use the BF16 base. No custom nodes are required.** Ordinary merged LoRA loaders can lose adapter updates through rounding. Update ComfyUI if the built-in node is unavailable.

| Version / 版本 | Download / 下载 | Size / 大小 | Steps / 步数 |
|---|---|---:|---:|
| v0.2.1 r256 | [r256 LoRA](qwen_image_2.1_viggle_turbo_v0.2.1_r256_comfy.safetensors) | 1.76 GB | 6 |
| v0.2.1 r128 | [r128 LoRA](qwen_image_2.1_viggle_turbo_v0.2.1_r128_comfy.safetensors) | 881 MB | 6 |
| v0.1 r64 · legacy / 旧版 | [r64 LoRA](Qwen-Image-2.1-viggle-turbo-4step-r64-comfyui-T8.safetensors) · [旧版说明 / legacy guide](README-v0.1.md) | 340 MB | 4 |

## 简体中文

1. 使用 **ComfyUI 0.37.0、前端 1.53.6 或更新版本**。选择一份 v0.2.1 LoRA，放入 `ComfyUI/models/loras/`。
2. 从 [Comfy-Org/Qwen-Image-2.1](https://huggingface.co/Comfy-Org/Qwen-Image-2.1) 下载三个底座文件，按下表放置。
3. 导入[文生图工作流](workflows/viggle_v021_text_to_image_ui.json)或[图片编辑工作流](workflows/viggle_v021_image_edit_ui.json)，选择模型；编辑时在 `LoadImage` 选择参考图。默认选用 r256，使用 r128 时切换 LoRA 文件即可。

| 文件 / File | 目录 / Folder |
|---|---|
| `qwen_image_2.1_bf16.safetensors` | `ComfyUI/models/diffusion_models/` |
| `qwen3vl_8b_int8_convrot.safetensors` | `ComfyUI/models/text_encoders/` |
| `qwen_image_2.1_vae_bf16.safetensors` | `ComfyUI/models/vae/` |

工作流使用 **Euler、六步、无 CFG**，按输出尺寸自动计算作者的 sigma 日程，并关闭 Qwen 前缀缓存。请保留调度组连线；仅把普通 KSampler 的步数改成 6 并不等价。另附[文生图 API](api_workflows/viggle_v021_text_to_image_api.json)和[编辑 API](api_workflows/viggle_v021_image_edit_api.json)。

支持 **BF16 DiT**。INT8 DiT 的融合 MLP 路径可能跳过 bypass hook；上表中的 INT8 **文本编码器**可以使用。使用原版 Qwen3-VL-8B，不使用 Zen student；不要叠加不同版本的蒸馏 LoRA，也不要挂在 Viggle 完整微调 Transformer 上。

转换保留全部 227 组源投影，通过块对角打包适配 ComfyUI 的融合 MLP；文件只含 LoRA，不含底座，没有再次截断 rank。已在禁用全部自定义节点的环境验证 512 像素文生图与编辑、完整权重映射和六步调度。内置 bypass 加载器仍标为实验性功能；复杂多参考图编辑并非均能准确遵循指令。

## English

These are **v0.2.1 six-step r256 / r128** adapters converted for native ComfyUI. The repository retains its historical `4step-r64` name; the old r64 file and [legacy instructions](README-v0.1.md) remain available.

Use **ComfyUI 0.37.0 / frontend 1.53.6 or newer**. Place one v0.2.1 adapter in `ComfyUI/models/loras/`, download the separate base files listed above, and open the [text-to-image](workflows/viggle_v021_text_to_image_ui.json) or [image-edit](workflows/viggle_v021_image_edit_ui.json) workflow. Select your model files and, for editing, a reference image. Switch the LoRA filename to use r128.

Load through **`LoraLoaderBypassModelOnly`, strength `1.0`, BF16 DiT**. The workflows use Euler, six steps, no CFG, the author's resolution-dependent sigma schedule, and disabled prefix caching. Keep the schedule group connected: setting an ordinary KSampler to six steps is not equivalent. INT8 DiT's fused MLP can skip bypass hooks; the separate INT8 Qwen3-VL-8B text encoder is supported. Do not use the Zen student encoder or stack distilled adapters/full fine-tunes.

All 227 source projection adapters are preserved through block-diagonal packing for ComfyUI's fused MLP. No base weights, further rank truncation, or weight merging are included. Native 512 px generation/editing, complete mapping, and schedule equivalence were validated with all custom nodes disabled. The built-in bypass loader is experimental; this is functional validation, not a comprehensive quality benchmark.

## Sources & license / 来源与许可

[Upstream model / 原模型](https://huggingface.co/Viggle/Qwen-Image-2.1-viggle-turbo) · [Conversion source / 转换源码](https://github.com/T8mars/comfyui-Zen-Image-Edit-T8/blob/main/scripts/convert_viggle_lora.py) · [SHA256SUMS](SHA256SUMS)

Weights retain the [Qwen Research License](LICENSE), for non-commercial research/evaluation; commercial use requires separate permission. / 权重沿用 Qwen Research License，仅限非商业研究与评估；商业使用须另行取得许可。v0.2.1 归属和修改见 [NOTICE-v0.2.1](NOTICE-v0.2.1) 与 [上游声明](NOTICE-UPSTREAM-v0.2.1)；旧版见 [NOTICE](NOTICE)。

## T8star

[B站 / Bilibili](https://space.bilibili.com/385085361) · [YouTube](https://www.youtube.com/@T8star-Aix/) · [Hugging Face](https://huggingface.co/t8star) · [API](https://api.seedance.nz/sign-up?aff=5f4w) · [免费画廊 / Gallery](https://www.openzhenzhen.com) · [在线 AI 应用 / AI apps](https://www.runninghub.ai/zh-cn/user-center/1907375370302308353/userPost?inviteCode=rh-v1121) · [ComfyUI 整合包 / Portable bundle](https://pan.quark.cn/s/264edb7e36bd)