> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bizyair.vip/llms.txt
> Use this file to discover all available pages before exploring further.

# Model Hub

> A collection of community-published and official models, ready to use without downloading

The Model Hub brings together community-published AI models and platform-official models. They can be mounted directly in ComfyUI nodes, and favorited, liked, and shared within the community. Think of it as an all-in-one marketplace: "model distribution + user upload & favorites + direct node mounting". Platform library models require no download, and community models can be reused with one click.

## Core Capabilities

| Capability                | Description                                                                                               |
| ------------------------- | --------------------------------------------------------------------------------------------------------- |
| Ready-to-use models       | Platform models mount directly to ComfyUI nodes — no local download needed                                |
| Multi-type classification | Multiple model types: `LoRA / Controlnet / Checkpoint / VAE / UNet / CLIP / Upscaler / Detection / Other` |
| Multiple base models      | A base-model ecosystem covering `Flux, SD, Illustrious, Wan Video, Qwen, Z-Image, LTX-2`, and more        |
| User uploads              | Users can upload their own models via the "Publish Model" entry and publish them to the community         |
| Edit & manage             | Edit model info and update versions from the detail page menu                                             |
| Like & share              | Supports likes and share-link generation so others can access the model                                   |
| Sort & filter             | Sort by Recent / Most used / Most favorited / Most liked; filter by type / base model / keyword           |

## Model Classification

Models in the library are organized along two dimensions: "type" and "base model".

### Model Types

| Type       | Description                                                          |
| ---------- | -------------------------------------------------------------------- |
| LoRA       | Low-rank fine-tuned models for injecting style / character / concept |
| Checkpoint | Full base model weights                                              |
| Controlnet | Conditioning control models for structure / pose guidance            |
| VAE        | Variational autoencoders for encoding and decoding                   |
| UNet       | Backbone network weights                                             |
| CLIP       | Text / vision encoders                                               |
| Upscaler   | Image upscaling models                                               |
| Detection  | Detection-type auxiliary models                                      |
| Other      | Other auxiliary models                                               |

## Uploading a Model

From the BizyAir top toolbar, click `Publish Model` to enter the publish-model flow.

### Publish Fields

| Field             | Description                                                                               |
| ----------------- | ----------------------------------------------------------------------------------------- |
| Model name        | Name of the model                                                                         |
| Model type        | One of nine types: LoRA / Checkpoint / Controlnet, etc.                                   |
| Version name      | Model version name (e.g., `v1`), to distinguish different versions of the same model      |
| Base model        | The base model it belongs to; choose "Other" if none applies                              |
| Cover image       | Upload a display image related to the model                                               |
| Model description | A brief introduction to the model                                                         |
| Public visibility | Whether to make it public; when enabled, other users can view it in the model marketplace |

### Publish Steps

1. Fill in the `Model name` and `Model type`.
2. Fill in the `Version name`, `Base model`, `Model description`, and other fields in turn.
3. Toggle `Public` to decide whether to publish it publicly.
4. Click or drag the model file into the upload area and wait for the upload to complete.
5. Click `Confirm` to publish the model.

## View Models

1. Click `Model Hub` in the top toolbar.
2. In Contents, you can view your uploaded public models and fork models from the model marketplace. In `Profile - My Published - Models`, you can view all models you have published.

## Using Models

1. Enter the ComfyUI workspace, select a ComfyUI node that loads a model, click the model name to open the model selection popup, select the desired model, and click Apply to use it in your ComfyUI workflow.
2. The model popup on the node will prioritize: Public models → My models → My favorites. `Public models` lets you select all public models from the community; `My models` lets you select models you have uploaded yourself; `My favorites` lets you select models you have favorited.

## Model Detail Page

Click a model cover in the model marketplace to open its detail page, which offers the following actions:

* View model details
* View artworks generated with this model
* **Like**: Like a model you enjoy; this affects its popularity ranking.
* **Favorite**: Favorite models that meet your needs for quick locating and selection when loading model nodes.
* **Share**: Generate a direct share code for others to access.
