CivArchive
    NebulaCanvas_FP8_FP16_T5_DEV - v2.0 UNET
    NSFW
    Preview 37934703
    Preview 37934778
    Preview 37934785
    Preview 37934783
    Preview 37934790
    Preview 37934789
    Preview 37934786
    Preview 37934809
    Preview 37937398
    Preview 37937449
    Preview 37937481
    Preview 37937576
    Preview 37938870
    Preview 37968299
    Preview 37968367
    Preview 37968496
    Preview 37969186
    Preview 37969562
    Preview 37979793
    Preview 37982799

    v2

    The updated version of the model is still in beta and currently includes about 85 styles (the list is given below). The ultimate goal is to add as many styles as possible and make them easily accessible to users, aiming to achieve a high level of realism. Compared to the previous version, this model has improved realism.

    To achieve the best results with styles, we recommend paying attention to a special prompting system developed by my friend Triple_Headed_Monkey (https://civarchive.com/user/Triple_Headed_Monkey). The archive attached under the model file shows an example of this prompt system in three versions: with a detailed description, a brief description and a description in natural language. You will get the best result when using the full version of prompt. Note that this is, in fact, a template indicating the details of the style and its features, and the prompt is entered in curly brackets {prompt}.

    That is, styles work in three ways, use them as it will be convenient for you.

    You can download examples of promptov as an archive in the attachment to the model. The archive contains a text file with examples of prompt in three types and a *.json file that you can use as additional styles in the MileHigh node https://civarchive.com/models/119246 ?modelVersionId=933860 by placing this file in the Data directory of this node and be sure to restart ComfyUI

    I hope you like this model, but I will continue to work on improving it.

    Chalk pastel mode

    Watercolor Silhouette mode

    v1

    "NebulaCanvas" is my first attempt to integrate styles from SDXL models into the the FLUX model. This model is in beta and may not be perfect yet, but it is trying to pave the way for a new dimension of FLUX image generation. In "NebulaCanvas", the emphasis is on artistic styles rather than realism.

    My friend Triple_Headed_Monkey and I redesigned the styles for the FLUX models. These styles are available through the ComfyUI_Mile High Style node. This will be a help for you in the direction of styles ComfyUI_MileHighStyler - Flux Mile High Beta | Flux Other | Civitai

    Description

    v2

    The updated version of the model is still in beta and currently includes about 85 styles (the list is given below). The ultimate goal is to add as many styles as possible and make them easily accessible to users, aiming to achieve a high level of realism. Compared to the previous version, this model has improved realism.

    To achieve the best results with styles, we recommend paying attention to a special prompting system developed by my friend Triple_Headed_Monkey (https://civitai.com/user/Triple_Headed_Monkey). The archive attached under the model file shows an example of this prompt system in three versions: with a detailed description, a brief description and a description in natural language. You will get the best result when using the full version of prompt. Note that this is, in fact, a template indicating the details of the style and its features, and the prompt is entered in curly brackets {prompt}.

    That is, styles work in three ways, use them as it will be convenient for you.

    You can download examples of promptov as an archive in the attachment to the model. The archive contains a text file with examples of prompt in three types and a *.json file that you can use as additional styles in the MileHigh node https://civitai.com/models/119246 ?modelVersionId=933860 by placing this file in the Data directory of this node and be sure to restart ComfyUI

    I hope you like this model, but I will continue to work on improving it.

    v1

    "NebulaCanvas" is my first attempt to integrate styles from SDXL models into the the FLUX model. This model is in beta and may not be perfect yet, but it is trying to pave the way for a new dimension of FLUX image generation. In "NebulaCanvas", the emphasis is on artistic styles rather than realism.

    My friend Triple_Headed_Monkey and I redesigned the styles for the FLUX models. These styles are available through the ComfyUI_Mile High Style node. This will be a help for you in the direction of styles ComfyUI_MileHighStyler - Flux Mile High Beta | Flux Other | Civitai

    Hide

    FAQ

    Checkpoint
    Flux.1 D

    Details

    Downloads
    412
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    11/2/2024
    Updated
    2/18/2026
    Deleted
    9/23/2025

    Files

    nebulacanvasFP8FP16T5_v20UNET.safetensors

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.