tinyLoader

Documentation

• Class name: ttN tinyLoader
• Category: 🌏 tinyterra/base
• Output node: False

The ttN tinyLoader node is designed for loading and initializing tiny models within the ComfyUI framework, facilitating the integration and management of lightweight models for efficient processing and analysis.

Input types

Required

• ckpt_name
  • Specifies the checkpoint name for loading the model, crucial for initializing the model with pre-trained weights.
  • Comfy dtype: COMBO[STRING]
  • Python dtype: str
• config_name
  • Determines the configuration name for the model, essential for setting up model parameters and behaviors.
  • Comfy dtype: COMBO[STRING]
  • Python dtype: str
• sampling
  • Defines the sampling strategy to be used by the model, affecting the generation or processing outcomes.
  • Comfy dtype: COMBO[STRING]
  • Python dtype: str
• zsnr
  • Specifies the zero-shot noise ratio, influencing the model's inference behavior.
  • Comfy dtype: BOOLEAN
  • Python dtype: float
• cfg_rescale_mult
  • Determines the CFG rescale multiplier, adjusting the model's configuration for specific tasks.
  • Comfy dtype: FLOAT
  • Python dtype: float
• vae_name
  • Specifies the VAE name for loading, essential for initializing the model with a specific variational autoencoder.
  • Comfy dtype: COMBO[STRING]
  • Python dtype: str
• clip_skip
  • Defines the number of layers to skip in the CLIP model, affecting the model's vision processing capabilities.
  • Comfy dtype: INT
  • Python dtype: int
• empty_latent_aspect
  • Specifies the aspect ratio for generating an empty latent space, crucial for initializing the model's latent representation.
  • Comfy dtype: COMBO[STRING]
  • Python dtype: str
• empty_latent_width
  • Determines the width of the empty latent space, essential for setting up the model's latent representation size.
  • Comfy dtype: INT
  • Python dtype: int
• empty_latent_height
  • Specifies the height of the empty latent space, crucial for defining the model's latent representation dimensions.
  • Comfy dtype: INT
  • Python dtype: int
• batch_size
  • Defines the batch size for processing, affecting the model's performance and resource utilization.
  • Comfy dtype: INT
  • Python dtype: int

Output types

• model
  • Comfy dtype: MODEL
  • The modified or initialized model ready for further processing or analysis within the ComfyUI framework.
  • Python dtype: str
• latent
  • Comfy dtype: LATENT
  • The generated or modified latent representation, ready for further processing or analysis.
  • Python dtype: str
• vae
  • Comfy dtype: VAE
  • The loaded or initialized VAE model, essential for variational autoencoder tasks.
  • Python dtype: str
• clip
  • Comfy dtype: CLIP
  • The loaded or initialized CLIP model, crucial for vision and language processing tasks.
  • Python dtype: str
• width
  • Comfy dtype: INT
  • The width of the generated or processed output, reflecting the dimensions of the model's output space.
  • Python dtype: int
• height
  • Comfy dtype: INT
  • The height of the generated or processed output, indicating the dimensions of the model's output space.
  • Python dtype: int

Usage tips

• Infra type: CPU
• Common nodes: unknown

Source code

class ttN_tinyLoader:
    version = '1.1.0'
    @classmethod
    def INPUT_TYPES(cls):
        aspect_ratios = ["width x height [custom]",
                        "512 x 512 [S] 1:1",
                        "768 x 768 [S] 1:1",
                        "910 x 910 [S] 1:1",

                        "512 x 682 [P] 3:4",
                        "512 x 768 [P] 2:3",
                        "512 x 910 [P] 9:16",

                        "682 x 512 [L] 4:3",
                        "768 x 512 [L] 3:2",
                        "910 x 512 [L] 16:9",

                        "1024 x 1024 [S] 1:1",                        
                        "512 x 1024 [P] 1:2",
                        "1024 x 512 [L] 2:1",

                        "640 x 1536 [P] 9:21",
                        "704 x 1472 [P] 9:19",
                        "768 x 1344 [P] 9:16",
                        "768 x 1216 [P] 5:8",
                        "832 x 1216 [P] 2:3",
                        "896 x 1152 [P] 3:4",

                        "1536 x 640 [L] 21:9",
                        "1472 x 704 [L] 19:9",
                        "1344 x 768 [L] 16:9",
                        "1216 x 768 [L] 8:5",
                        "1216 x 832 [L] 3:2",
                        "1152 x 896 [L] 4:3",
                        ]

        return {"required": { 
                        "ckpt_name": (folder_paths.get_filename_list("checkpoints"), ),
                        "config_name": (["Default",] + folder_paths.get_filename_list("configs"), {"default": "Default"} ),
                        "sampling": (["Default", "eps", "v_prediction", "lcm", "x0"], {"default": "Default"}),
                        "zsnr": ("BOOLEAN", {"default": False}),
                        "cfg_rescale_mult": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}),

                        "vae_name": (["Baked VAE"] + folder_paths.get_filename_list("vae"),),
                        "clip_skip": ("INT", {"default": -1, "min": -24, "max": 0, "step": 1}),

                        "empty_latent_aspect": (aspect_ratios, {"default":"512 x 512 [S] 1:1"}),
                        "empty_latent_width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                        "empty_latent_height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                        "batch_size": ("INT", {"default": 1, "min": 1, "max": 64}),
                        },
                "hidden": {"prompt": "PROMPT", "ttNnodeVersion": ttN_tinyLoader.version, "my_unique_id": "UNIQUE_ID",}
                }

    RETURN_TYPES = ("MODEL", "LATENT", "VAE", "CLIP", "INT", "INT",)
    RETURN_NAMES = ("model", "latent", "vae", "clip", "width", "height",)

    FUNCTION = "miniloader"
    CATEGORY = "🌏 tinyterra/base"

    def miniloader(self, ckpt_name, config_name, sampling, zsnr, cfg_rescale_mult, vae_name, clip_skip,
                       empty_latent_aspect, empty_latent_width, empty_latent_height, batch_size,
                       prompt=None, my_unique_id=None):

        model: ModelPatcher | None = None
        clip: CLIP | None = None
        vae: VAE | None = None

        # Create Empty Latent
        latent = sampler.emptyLatent(empty_latent_aspect, batch_size, empty_latent_width, empty_latent_height)
        samples = {"samples": latent}

        model, clip, vae = loader.load_checkpoint(ckpt_name, config_name, clip_skip)
        if vae_name != "Baked VAE":
            vae = loader.load_vae(vae_name)

        if sampling != "Default":
            MSD = comfy_extras.nodes_model_advanced.ModelSamplingDiscrete()
            model = MSD.patch(model, sampling, zsnr)[0]

        if cfg_rescale_mult > 0:
            CFGR = comfy_extras.nodes_model_advanced.RescaleCFG()
            model = CFGR.patch(model, cfg_rescale_mult)[0]

        return (model, samples, vae, clip, empty_latent_width, empty_latent_height)