Qwen Image 2.1 Text Encoder

The original text encoder of Qwen Image 2.1 (Qwen3-VL 8B), unchanged, so you can download it from Civitai.

What it is

• The model that reads your prompt for Qwen Image 2.1: Qwen3-VL 8B, packaged by Comfy-Org for ComfyUI

• Two files, byte-identical to Comfy-Org's

  • int8 convrot, 9.35 GB, the default

  • w4a8, 6.3 GB, for 8 GB cards and low RAM

• Uncensored out of the box

• Same file names as in ComfyUI's official templates, so they load as is

No abliteration needed

Qwen Image 2.1 renders nudity and adult scenes with this original encoder. Noct Q uses it.

Works with

• Qwen Image 2.1, base and turbo, for text-to-image and editing

• Noct Q, the uncensored Qwen Image 2.1 in Turbo and Base versions: https://civarchive.com/models/2958896

• The Qwen Image 2.1 VAE: https://civarchive.com/models/2960925

• Not with Krea 2, which uses the 4B Qwen3-VL encoder

Install

• Put it in ComfyUI/models/text_encoders/ and load it with the Load CLIP node, type qwen_image (ComfyUI 0.37.0 or newer)

About the images

Rendered with Qwen Image 2.1 Base (int8 convrot, 25 steps, cfg 1, 1024x1536) and this encoder. Each prompt tests one thing the encoder has to get right: text in English and Chinese, counts, placement, two people with their own clothes.

中文：Qwen Image 2.1 原版文本编码器（Qwen3-VL 8B），未经修改，与 Comfy-Org 发布的文件逐字节一致：int8 convrot（9.35 GB，默认）和 w4a8（6.3 GB，适合 8G 显存）。放入 ComfyUI/models/text_encoders/，用 Load CLIP 节点（type 选 qwen_image）加载。Qwen Image 2.1 使用原版编码器即可生成成人内容，无需 abliterated 版本。仅限非商业用途。

License

Built with Qwen. Qwen is licensed under the Qwen RESEARCH LICENSE AGREEMENT, Copyright (c) 2026 Hangzhou Tongyi Laboratory Technology Co., Ltd. All Rights Reserved. Free, non-commercial use only. LICENSE and NOTICE are in the zip.