Nodes/Diztraido Nodes/Load Flux.1 Models + LoRAs
ComfyUI Node

Load Flux.1 Models + LoRAs

The Flux.1 triple-loader, plus a 16-slot LoRA stack in the same node

By jadervasque·Created 2 months ago·Updated 2 months ago· 0
Load Flux.1 Models + LoRAs
    • model
    • clip
    • vae
    unet_name
    weight_dtype
    clip_name1
    clip_name2
    type
    vae_name
    lora_count0
    lora_1
    strength_model_11.00
    strength_clip_10.00
    lora_2
    strength_model_21.00
    strength_clip_20.00
    lora_3
    strength_model_31.00
    strength_clip_30.00
    lora_4
    strength_model_41.00
    strength_clip_40.00
    lora_5
    strength_model_51.00
    strength_clip_50.00
    lora_6
    strength_model_61.00
    strength_clip_60.00
    lora_7
    strength_model_71.00
    strength_clip_70.00
    lora_8
    strength_model_81.00
    strength_clip_80.00
    lora_9
    strength_model_91.00
    strength_clip_90.00
    lora_10
    strength_model_101.00
    strength_clip_100.00
    lora_11
    strength_model_111.00
    strength_clip_110.00
    lora_12
    strength_model_121.00
    strength_clip_120.00
    lora_13
    strength_model_131.00
    strength_clip_130.00
    lora_14
    strength_model_141.00
    strength_clip_140.00
    lora_15
    strength_model_151.00
    strength_clip_150.00
    lora_16
    strength_model_161.00
    strength_clip_160.00

    Load Flux.1 Models + LoRAs is the Diztraido Flux.1 loader from its non-LoRA sibling, with up to 16 LoRA slots welded onto the same node. Instead of loading the model with one node, then dragging a chain of LoraLoader nodes across your canvas and wiring strengths through each one, you set lora_count, pick files, dial in strengths, and get one clean model/clip/vae bundle out the other end. The composite nodes in this pack are all about shrinking the wire spaghetti of a stock Flux graph; this is the most aggressive version of that, because LoRA stacking is normally where the graph gets ugly.

    It's a real quality-of-life win if you regularly stack several LoRAs - a style LoRA plus a character LoRA plus a resolution or detail one, say. It's overkill if you never use LoRAs at all; get the plain loader instead.

    How it works

    Mechanically it's the same wrapper as DiztraidoLoadFlux1Models: it merges the required inputs of ComfyUI's UNETLoader, DualCLIPLoader (type defaulted to flux), and VAELoader into one panel, then appends a LoRA section. Each of the 16 slots is a lora_name picker plus strength_model and strength_clip. At execution it loads the base model, then applies the active LoRAs sequentially through ComfyUI's LoraLoader - the output of one feeds the input of the next - so the order in the slots is the order of application. Slots with no file selected or both strengths at zero are skipped, and lora_count controls how many slots are live, which keeps the widget panel from being a wall of empty pickers.

    The inputs that matter

    The model side is identical to the plain loader: unet_name, weight_dtype (fp8_e4m3fn is the sensible default on consumer cards), clip_name1 / clip_name2 (t5xxl + clip_l for Flux.1), type (leave on flux), vae_name (pixel_space).

    Then the LoRA half:

    • lora_count - 0 to 16, how many slots are active. This is the one you set first.
    • lora_1lora_16 - the file pickers, from models/loras/.
    • strength_model_1 and strength_clip_1 - defaults of 1.0 and 0.0 respectively. For Flux, strength_clip usually wants to be 0 anyway (Flux.1 LoRAs are almost always model-only), and the default agrees.

    Outputs: model, clip, vae - ready to wire into text encoding and sampling.

    How to install

    Part of Diztraido Nodes; you get it with the whole pack. ComfyUI Manager → search "Diztraido Nodes" → install diztraido-nodes → restart. Or:

    cd ComfyUI/custom_nodes
    git clone https://github.com/jadervasque/ComyUI-Diztraido.git
    

    Restart, and it's under the Diztraido flux category. No extra dependencies, no bundled models - your base model, encoders, and LoRAs all need to be in your existing ComfyUI/models/ tree.

    Common issues

    Two things worth knowing. First, ordering matters: LoRAs are applied top to bottom, and with styles especially, order changes the result. Second, a strength of 1.0 on a stack of three or four LoRAs will frequently be too much - overlapping styles is how you get that melted, oversaturated look. Start around 0.6–0.8 per LoRA when stacking, and remember that LoRA strengths past 1.0 exist but usually only hurt. Also, if a lora_name dropdown is empty, your LoRA file isn't in models/loras/ - the node lists what's actually on disk.

    CategoryDiztraido/flux

    Inputs (55)

    NameTypeDefaultDescription
    unet_nameCOMBO0 options:
    weight_dtypeCOMBO4 options: default, fp8_e4m3fn, fp8_e4m3fn_fast, fp8_e5m2
    clip_name1COMBO0 options:
    clip_name2COMBO0 options:
    typeCOMBO12 options: sdxl, sd3, flux, hunyuan_video, hidream, hunyuan_image, +6
    vae_nameCOMBO1 options: pixel_space
    lora_countINT00–16
    lora_1COMBOThe name of the LoRA.
    strength_model_1FLOAT1.00-100–100
    strength_clip_1FLOAT0.00-100–100
    lora_2COMBOThe name of the LoRA.
    strength_model_2FLOAT1.00-100–100
    strength_clip_2FLOAT0.00-100–100
    lora_3COMBOThe name of the LoRA.
    strength_model_3FLOAT1.00-100–100
    strength_clip_3FLOAT0.00-100–100
    lora_4COMBOThe name of the LoRA.
    strength_model_4FLOAT1.00-100–100
    strength_clip_4FLOAT0.00-100–100
    lora_5COMBOThe name of the LoRA.
    strength_model_5FLOAT1.00-100–100
    strength_clip_5FLOAT0.00-100–100
    lora_6COMBOThe name of the LoRA.
    strength_model_6FLOAT1.00-100–100
    strength_clip_6FLOAT0.00-100–100
    lora_7COMBOThe name of the LoRA.
    strength_model_7FLOAT1.00-100–100
    strength_clip_7FLOAT0.00-100–100
    lora_8COMBOThe name of the LoRA.
    strength_model_8FLOAT1.00-100–100
    strength_clip_8FLOAT0.00-100–100
    lora_9COMBOThe name of the LoRA.
    strength_model_9FLOAT1.00-100–100
    strength_clip_9FLOAT0.00-100–100
    lora_10COMBOThe name of the LoRA.
    strength_model_10FLOAT1.00-100–100
    strength_clip_10FLOAT0.00-100–100
    lora_11COMBOThe name of the LoRA.
    strength_model_11FLOAT1.00-100–100
    strength_clip_11FLOAT0.00-100–100
    lora_12COMBOThe name of the LoRA.
    strength_model_12FLOAT1.00-100–100
    strength_clip_12FLOAT0.00-100–100
    lora_13COMBOThe name of the LoRA.
    strength_model_13FLOAT1.00-100–100
    strength_clip_13FLOAT0.00-100–100
    lora_14COMBOThe name of the LoRA.
    strength_model_14FLOAT1.00-100–100
    strength_clip_14FLOAT0.00-100–100
    lora_15COMBOThe name of the LoRA.
    strength_model_15FLOAT1.00-100–100
    strength_clip_15FLOAT0.00-100–100
    lora_16COMBOThe name of the LoRA.
    strength_model_16FLOAT1.00-100–100
    strength_clip_16FLOAT0.00-100–100

    Outputs (3)

    NameTypeDescription
    modelMODEL
    clipCLIP
    vaeVAE