CivArchive
    Preview 149293064
    Preview 149293051
    Preview 149293065
    Preview 149293069
    Preview 149293068
    Preview 149293072
    Preview 149293071
    Preview 149293074
    Preview 149293073
    Preview 149293270
    Preview 149293268
    Preview 149293273
    Preview 149293276
    Preview 149293279
    Preview 149293290
    Preview 149293301
    Preview 149293295
    Preview 149293302
    Preview 149293342
    Preview 149293343

    Main characteristics of this LoRA are as follows:

    · Reversal film / slide film style, with strong contrast and rich, intense color saturation.

    · Improved lighting and shadow effects, especially extremely powerful and expressive sunlight rendering.

    · Based on personal aesthetic preference, female characters tend toward large breasts, slim waist, and wide hips.

    Important: This LoRA has a very strong style with distinct personal preferences. In other words, it cannot guarantee diversity in body types etc. (You can achieve more variety by using detailed and specific prompts.)

    Info for MiniMax-H3-Image

    Note: This LoRA is designed for MiniMax-H3-Image (T2I model), NOT MiniMax-H3 (video model).

    Key Improvements:

    • Improved lighting and shadow effects.

    • More attractive female characters.

    • Reduced the coarse graininess and jagged edges present in the original model.

    Recommended Sampler: euler_simple
    (This LoRA was trained using euler_simple. Using other samplers may reduce its effectiveness.)

    Recommended Strength: 1.0–1.25

    Important Tip: To reduce the roughness of the original model's output, it is strongly recommended to connect a blank white image to the MiniMax H3 Image Conditioning node.

    The white image can be relatively small. For example, for a 9:16 aspect ratio, a resolution of 512 × 910 is sufficient.

    Recommended Resolution: Generate images at 16 megapixels for the best possible image quality, then downscale them to your desired resolution as needed.

    Strongly recommended: Use the ComfyUI-H3-Image-EasyFix node (GitHub) to further reduce coarse graininess and improve image quality.

    When using it together with this LoRA, it is recommended to set the MiniMax H3 Image Steering node strength to 0.3–0.5.

    Info for MOSS-Image-Base v1.0

    This LoRA was trained on MOSS-Image-base and can be used with lodestones/MOSS-Image-Base-8-steps-dmd. In theory, it should also work with the SFT model, but this has not been tested.

    • Recommended Settings:

      • Strength: 1

      • Sampler: euler_sgm_uniform

    Info for Flux.3 Klein v1.0

    This Lora supports both base and step distilled model. NSFW supported.

    • Recommended Setting:

      • strength: 1

      • sampler: euler_simple

    License NOTE:

    This LoRA is released under the FLUX Kommunity License v.1.0: non-commercial use, plus commercial output generation by Qualifying Users with less than US$5M in gross annualized revenue, including affiliates, subject to the license terms. For other uses, see commercial licensing.

    Info for QwenImage 2.1 v1.0

    Recommended strength: 1.0

    Recommended sampler & scheduler: euler_simple

    NOTE:

    License

    This LoRA is a derivative work based on Qwen-Image-2.1. It is released under the Qwen Research License Agreement. Commercial use is not permitted without the appropriate commercial license from Qwen. Please refer to the original Qwen-Image-2.1 license: https://huggingface.co/Qwen/Qwen-Image-2.1/blob/main/LICENSE

    Info for Krea2 Turbo v1.0

    Recommended strength: 0.5–1.0, 0.75 is the best in my test.

    Recommended sampler & scheduler: er_sde_simple

    Info for ERNIE-image (and turbo) v1.1

    This LoRA is compatible with ERNIE-image and its distilled model ERNIE-image-turbo.

    • For the ERNIE-image-turbo model, it can be used directly.

    • In addition, this LoRA can significantly reduce the yellowish cast and rough artifacts of the ERNIE-image-turbo model, producing images with better white balance and finer, more detailed quality.

    Recommended strength: 1.0

    Info for z-image-turbo v2.0

    This is an improved version of v1.0, with the main changes as follows:

    · Overall tone is brighter, with more details preserved in the shadows

    · Reduced some unreasonable light and shadow effects

    · Contrast is not as strong/intense as in v1.0

    Rather than calling it an “upgrade,” it’s more accurate to say it’s a variant with a different color tone and style. Users can choose between the two versions based on their personal preference.

    Recommended strength: 0.7–1.0

    Recommended sampler & scheduler: euler_capitanZIT (https://github.com/capitan01R/ComfyUI-CapitanZiT-Scheduler)

    Info for z-image-turbo v1.0

    Recommended strength: 0.5–1.0

    Recommended sampler & scheduler: euler_capitanZIT (https://github.com/capitan01R/ComfyUI-CapitanZiT-Scheduler)

    Info for z-image v1.0

    This is a version retrained based on the z-image model.

    Recommended strength: 1.0

    Recommended sampler & scheduler: res_multistep beta 25steps

    Description

    Note: This LoRA is designed for MiniMax-H3-Image (T2I model), NOT MiniMax-H3 (video model).

    Key Improvements:

    • Improved lighting and shadow effects.

    • More attractive female characters.

    • Reduced the coarse graininess and jagged edges present in the original model.

    Recommended Sampler: euler_simple
    (This LoRA was trained using euler_simple. Using other samplers may reduce its effectiveness.)

    Recommended Strength: 1.0–1.25

    Important Tip: To reduce the roughness of the original model's output, it is strongly recommended to connect a blank white image to the MiniMax H3 Image Conditioning node.

    The white image can be relatively small. For example, for a 9:16 aspect ratio, a resolution of 512 × 910 is sufficient.

    Recommended Resolution: Generate images at 16 megapixels for the best possible image quality, then downscale them to your desired resolution as needed.

    Strongly recommended: Use the ComfyUI-H3-Image-EasyFix node (GitHub) to further reduce coarse graininess and improve image quality.

    When using it together with this LoRA, it is recommended to set the MiniMax H3 Image Steering node strength to 0.3–0.5.

    FAQ

    LORA
    MiniMax H3

    Details

    Downloads
    25
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/17/2026
    Updated
    11/20/2026
    Deleted
    -

    Files

    RadianceChromeVoluptuous_MiniMax-H3-Image_v1.0.safetensors