CivArchive
    Text to Image with Qwen image 2.1 - NSFW Workflow - v1.0
    NSFW
    Preview 145852571
    Preview 145754053
    Preview 145754055
    Preview 145754046

    Qwen Image 2.1 - NSFW Workflow (ComfyUI)

    ❗You are responsible for Civitai's terms and the laws where you live. Do not use it to depict real people.❗

    Complete ready-to-use ComfyUI workflow for Qwen Image 2.1 focused on high-quality NSFW / erotic generations.

    What's included Optimized load setup for Qwen Image 2.1 (UNET + CLIP + VAE) Stacked LoRAs for NSFW anatomy & posture, skin detail / micro-texture, lingerie & sheer fabrics, and overall aesthetic enhancement Clean positive + negative prompt structure Resolution selector (default 16:9 widescreen) KSampler settings tuned for Qwen (50 steps, CFG 2.5, er_sde + beta scheduler) Save Image node with proper naming

    How to use

    1. Type your idea in the green Prompt box (short description is enough)

    2. Toggle Enhance if you want the prompt expanded into a detailed version

    3. Enable the LoRAs you need (or leave the recommended ones on)

    4. Set resolution

    5. Hit Run

    Recommended default settings Steps: 50 CFG: 2.5 Sampler: er_sde Scheduler: beta

    Notes Designed for realistic / semi-realistic NSFW outputs Works best with detailed clothing, lighting, and body descriptions All LoRA strengths are pre-tuned — adjust if needed

    Feel free to share your results!

    Description

    FAQ

    Workflows
    Qwen 2.1

    Details

    Downloads
    272
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/15/2026
    Updated
    10/18/2026
    Deleted
    -

    Files

    textToImageWithQwenImage2_v10.json

    Mirrors

    textToImageWithQwenImage2_v10.zip

    Mirrors

    textToImageWithQwenImage2_v10.zip

    Mirrors