CivArchive
    Animax - Anima Finetune Model - v0.1
    NSFW
    Preview 122081507
    Preview 122077181
    Preview 122076659
    Preview 122085260
    Preview 122079890

    Anima

    Go to the official Anima Huggingface repo for answers.

    You need to use qwen_3_06b_base.safetensors for text encoder, and qwen_image_vae.safetensors for VAE.

    Installing and running

    The model is natively supported in ComfyUI. The above image contains a workflow; you can open it in ComfyUI or drag-and-drop to get the workflow. The model files go in their respective folders inside your model directory:

    • animax.safetensors goes in ComfyUI/models/diffusion_models

    • qwen_3_06b_base.safetensors goes in ComfyUI/models/text_encoders

    • qwen_image_vae.safetensors goes in ComfyUI/models/vae (this is the Qwen-Image VAE, you might already have it)

    Generation settings

    • The preview version should be used at about 1MP resolution. E.g. 1024x1024, 896x1152, 1152x896, etc.

    • 30-50 steps, CFG 4-5.

    • A variety of samplers work. Some of my favorites:

      • er_sde: neutral style, flat colors, sharp lines. I use this as a reasonable default.

      • euler_a: Softer, thinner lines. Can sometimes tend towards a 2.5D look. CFG can be pushed a bit higher than other samplers without burning the image.

      • dpmpp_2m_sde_gpu: similar in style to er_sde but can produce more variety and be more "creative". Depending on the prompt it can get too wild sometimes.

    Description

    FAQ

    Comments (9)

    DazrockFeb 23, 2026· 1 reaction
    CivitAI

    Feel more sample images?
    Good job btw!

    rityak
    Author
    Feb 24, 2026· 2 reactions

    Thank you, there will be more examples in the next version, which is already in progress.

    seawolf338Feb 24, 2026· 3 reactions
    CivitAI

    Can you give more information about this merge? What was merged with what?


    rityak
    Author
    Feb 24, 2026· 3 reactions

    This is my trained model, the original file is https://huggingface.co/rityak/animax/tree/main.

    The training result was transferred to a mixture of the original Anima and AnimaYume models.

    Anima and AnimaYume were mixed using the crossover method.

    MasterArtFeb 25, 2026· 2 reactions
    CivitAI

    More stable than original one and default style is great. Thank you!

    GPUPoorChadFeb 27, 2026· 2 reactions
    CivitAI

    Overall better quality, but it might be a bit worse at prompt adherence and pussy anatomy occasionally

    rityak
    Author
    Feb 27, 2026

    Yes, this problem exists. I suspect its related to some of the images in the dataset. Perhaps expanding the dataset will improve the situation. Im currently working on it.

    GPUPoorChadFeb 27, 2026

    @rityak 👀👀

    cococolacakeMar 5, 2026· 1 reaction
    CivitAI

    noticeably more stable than the base! works great!

    Checkpoint
    Anima

    Details

    Downloads
    539
    Platform
    CivitAI
    Platform Status
    Available
    Created
    2/23/2026
    Updated
    4/30/2026
    Deleted
    -