CivArchive
    [Qwen Image 2.1] Character Design Sheet Maker (Workflow) - Qwen Image 2.1 [V3.0]
    NSFW
    Preview 144458015
    Preview 144453459
    Preview 144453477
    Preview 144453465
    Preview 144453485
    Preview 144453461

    Changelog

    V3.0 [FINAL FOR QWEN IMAGE 2.1]

    This will be my final update for the Qwen Image 2.1 lineup. There is little reason to add or tweak further, so I am officially handing the torch over to the community.

    The major concept shift in this release is native support for static prompts. I've seen many comments asking: "Why use a PE (Prompt Enhancer) when you can just send a static prompt?" To an extent, that's completely valid - if you're not a writer or artist and simply want to quickly animate your character, static prompts are the way to go.

    However, I kept the Auto-PE mode for users who still need granular control over character traits, outfit consistency, and fine details.

    PE Model Revamp

    A major change in this version is the overhaul of the PE pipeline itself. For many, GGUF proved to be too slow, while running Qwen 3.5 9B natively was too VRAM-heavy. I decided to take a different approach and switch to Qwen 3.5 4B Safetensors.

    Why this model?

    • Tests showed that Qwen 3 VL (4B/8B) struggled with complex instruction following, frequently repeating the system prompt or outputting low-quality sheets.

    • Switching to Qwen 3.5 4B eliminated the repetition issue, though initial output lacked detail.

    • I fixed this by streamlining the system prompt. Enabling thinking mode produced noticeably better results: the model reasons through scene composition step-by-step (though it takes a bit more generation time).

    Ultimately, prompt generation now takes between 30 seconds and 2 minutes (for int8), with much higher reproducibility and fewer project dependencies.

    Two Workflows in One Bundle

    I combined V1 and V2 into a single unified package containing two specialized workflows:

    1. ..._Production - Use this if you want a polished, highly detailed, and aesthetically rich Character Sheet.

    2. ..._Simple - Use this if you want a streamlined, production-ready Character Sheet built strictly for video animation.

    Usage Recommendations

    1. PE Model Selection: You can use other PE models (local or cloud-based). However, for local deployment, I highly recommend Qwen 3.5 4B / 9B depending on your available VRAM. On an 8GB VRAM setup using Qwen 3.5 4B int8, I get around 30 tokens/sec.

    2. Static Prompts: If you don't need manual control over character backstory or precise details, stick to static prompts to speed up generation significantly.

    Mode PE Static Prompt Production + - Simple + +

    1. Performance Tip: If PE token generation speed drops unexpectedly, try re-queuing the generation. This appears to be a caching issue, and restarting restores normal speed.

    Full Changelog (V3.0)

    • Dropped GGUF support in favor of native Safetensors.

    • Rewrote system prompt to optimize instruction adherence for smaller LLM sizes.

    • Merged V1 (Production) and V2 (Simple) workflows into a single release package.

    Resource Links


    V2.0

    In version 2.0, based on community feedback, I redesigned the workflow to be much more task-focused: minimal unnecessary typography, maximum focus on key character elements, body structure, and facial details. This layout makes it drastically easier for video generation models to read character details and maintain consistency.

    How It Works

    The PE (Prompt Enhancer) generates 2 to 3 main panels:

    1. Facial Expressions

    2. Poses

    3. Equipment & Accessories (optional)

    This amount of information is ideal for video models to produce coherent, high-quality results.

    Layout Rules & Features:

    1. Faceless Entities: If the object or character does not have a human face (e.g., mask, robot, inanimate object), only a single neutral close-up panel is generated.

    2. Expression Variety: The set of emotional expressions adapts to the character's backstory and description, but a neutral expression always comes first.

    3. Gear & Accessories: If accessories or gear are present, a separate 3rd panel displays them in detail, isolated from the main character.

    4. character_description Parameter: Helps the PE better understand character traits and posture. For instance, providing a description like "Ayaka is a slender, fragile girl. She is mostly shy, but occasionally gives a subtle smile" allows the PE to choose matching expressions and poses. This parameter is optional (the PE can infer details strictly from the image), but manually specifying details yields much higher accuracy.

    Important Considerations:

    1. Input Image: The workflow performs best when the reference image shows a full-body character facing forward. The PE is intentionally constrained from hallucinating missing body parts unless explicitly specified in the character_description. This prevents unwanted inconsistencies like altered height, age, or outfit details.

    2. Two Workflow Variants: They differ only in how the PE prompt is generated:

      • For GPUs with <12GB VRAM: Use the standard version (without the _native_pe suffix).

      • The official Generate Text node is currently unoptimized for Qwen 3.5 9B, which can cause prompt generation on lower-end hardware to exceed 30 minutes. The developers are aware of this issue and working on performance fixes.

    Summary of Changes (v2.0):

    • Complete Redesign: Shifted focus from purely stylistic/artistic design sheets to a functional, production-ready Character Design Sheet optimized for video models.

    • Updated Text Field: Replaced entity_name (name only) with character_description (full personality & detail specifications).

    • Native PE Support: Added a secondary workflow utilizing the official PE execution method.

    • Usability: Added clear explanatory comments inside the ComfyUI node graphs.


    About this Workflow

    This workflow allows you to generate detailed Character Design Sheets from a single reference image. These sheets can later be used as reference frameworks for video generation models like Seedance 2.5 or MiniMax H3.

    In my experience, Qwen Image 2.1 is the first open-source image editing model capable of natively generating clean, highly detailed character design sheets right out of the box.

    • CLIP Encoder: Use FP32 / FP16 / BF16 precision for maximum image quality and prompt adherence. Tests show that INT8 and other lower-bit quantizations introduce noticeable artifacts and reduce image clarity. Don't skimp on SSD space — install qwen3vl_8b_bf16.safetensors.

    • Sampler & Scheduler: Use res_2m + beta to reduce artifacts and improve detail. While I haven't run extensive benchmarks on every combination, this pairing yielded the best results.

    • Resolution (Megapixels):

      • 3.4 MP — Sweet spot for speed and clarity.

      • 6.0 MP — Superior quality and detail, though generation time increases significantly.

    How to Use

    1. Install custom nodes via ComfyUI Manager:

    2. Download Qwen PE I2I or Qwen 3.5 4B/9B weights and place them into ComfyUI/models/text_encoders/.

    3. Run generation: Load your reference image, select the path to the PE model, and start generating.

    Useful Resources

    Description

    • Dropped GGUF support in favor of native Safetensors.

    • Rewrote system prompt to optimize instruction adherence for smaller LLM sizes.

    • Merged V1 (Production) and V2 (Simple) workflows into a single release package.

    FAQ

    Comments (8)

    chrisbraeuer41172035Oct 7, 2026
    CivitAI

    I think it would be more usefull if it can put up a face close up instead of the black white shots. You get the outlines already with the 3 perspective view. Just my opinion.

    NeuroContent
    Author
    Oct 7, 2026

    Doesn't the finished image already have a panel with the character's face (or rather, with various facial expressions)? If that's not what you mean, please explain in more detail.

    I mean the sheet overall looks great. but i am not sure if that is the best approach. I would rather have a simplified version. with a really close up face and the 3 full body shots. nothing else. no silhouettes, props and expressions. I don´t think they do much but cost pixel space. Yes the faces are there but i think its better to have one big high rez face than 3 small faces, 3 sihloutettes and lots of micro prop shots. RN i use 1024x2048 with x height. Upper part is the face close up and the lower part the 3 perspectives. So i can fit 2 chars in a full 2048x2048 budget. The extras like props silhouettes and expressions courd go on a 2nd page also 1024x2048. So if using more chars one could choose to take only the minimal ref ( big face and 3 perspectives) at half the budget or the full reference with everything at full buddget. Most of the time you will only need the minimal one. And since the sice affects generaion time a simplified reference would be cool.

    NeuroContent
    Author
    Oct 9, 2026

    @chrisbraeuer41172035 Overall, this makes sense if your goal is simply to animate a character. This approach does save space on the page and is perhaps easier on the eyes.

    In my implementation, I still tried to think like an artist or screenwriter. If my film has a character whose facial expressions are severely limited (for example, a perpetually sad person in a noir film), I would create a neutral, sad emotion for them, as well as an emotion with a subtle smile. Everything is tailored to the character's personality I want to convey). This approach gives me explicit control over the character's state throughout my generated video, since the video generative model no longer invents an emotion that might not suit me.

    In any case, your solution is no worse, and I support this simplification. I think even when it comes to character sheet generation, everyone will find their own approach to implementation.😀

    @NeuroContent It for sure looks great. From a presentation standpoint.

    Tr0canterOct 9, 2026
    CivitAI

    How can we add or change "things" to the final prompt? IE my son has long hair and the model is assuming that he is a girl/woman so I get the character sheet but he has breasts LOL

    NeuroContent
    Author
    Oct 9, 2026

    If you're using the Production workflow, you can combine the "Entity Name" node with the "Text (Multiline)" node using "Join String." You can already specify your character's description in "Text (Multiline)."

    If you are using the Simple workflow, there is already a "Character Description" node, where you can also specify a description of the character.

    Yes, I didn't take into account the character description approach from Simple to Production in this version. I'll try to release a fix.

    axalaOct 9, 2026· 1 reaction
    CivitAI

    I cant find redqw21UNLOCKEDV2TI2I_redqw21UNLOCKEDV2_int8.safetensors

    Workflows
    Qwen 2.1

    Details

    Downloads
    3,203
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/2/2026
    Updated
    11/22/2026
    Deleted
    -

    Files

    Character_Sheet.zip

    Mirrors

    CivitAI (1 mirrors)