# Generate Anything Quickstart The Generate Anything service is currently in an early access stage, so whilst Generate Anything is in these earlier stages, please bear with us!\ \ Before anything else, you need to started by opening Animate Anything website and signing in or signing up.

Next, you have to decide if you want to create a model using either our Image to 3D tool, or out Text to 3D tool. To begin with, we'll take a look at Image to 3D, but go ahead and scroll ahead if you're looking for Text to 3D instead. ### **Image to 3D Model Generation** Firstly, navigate to Generate Anything, which can be found on the top right of the Anything World toolbar.

On the Generate Anything homepage, select "Create from image" (right-hand button)

Either Drag and drop your image file or click "Browse your files" to navigate to the image file you want to use. Remember to refer to our Best Practices for Best Results page before uploading so you can properly prepare your image for the best outcome. You can then add a Model Name in the provided text box, however this is optional.\ \ The mandatory checkbox confirms you have read and accepted our system constraints, as well as the terms of use of the tool. {% hint style="warning" %} **Image-To-3D constraints are as follows:** * Use a transparent background or a single-coloured one. * Feature just one object, creature or character in the image. You should avoid even floor elements, and shadows on the floor or background. * The image should be moderately illuminated. Avoid excessive occlusion or extreme shininess. * Creatures or characters in images should display all their limbs clearly and follow conventional poses, ideally symmetrical. Otherwise, it is more likely to get an unexpected number of limbs in the model or other defects. * The image should be of high quality, avoiding blurriness, noise, text and other watermarks. Even in those conditions, please note that the system is not perfect and fine-grained details such as fingers, face details or accessories may not have outstanding quality. * Use JPG, JPEG or PNG images. If you are prompted that the image exceeds the file size limit, try reducing the resolution or modifying compression settings. {% endhint %} When you're ready to proceed, press Generate.

A properly filled out submission of image to 3D Model showing the image uploaded, the model name, the system constraints and terms of use accepted, ready for generation.

After pressing Generate, you will be taken to the generation progress screen, which shows you the status of your generation. This process can take up to six minutes.

An Image to 3D Model Generation in progress

Once your model has finished processing, you will be presented with the finished generation.

A generated model of a giraffe created using an image

From here, you can choose to download the model as-is, generate another model, or go ahead and animate the model using our Animate Anything tool. Your generated model will also be avaliable to view in My World, under *generated models*, so you can access it any time! ### **Text to 3D Model Generation** If you haven't already, navigate to Generate Anything, which can be found on the top right of the Anything World toolbar.

On the Generate Anything homepage, select "Create from Text prompt" (left-hand button)

Write your text prompt into the provided text box, then select the appropriate checkboxes. The **Optional** checkbox provides an option to generate your model with a standard pose that would be best suited for animation, should you want to use this. If you have a particular pose in mind for your prompt, it is recommended that you keep this *unchecked.* The mandatory checkbox confirms you have read and accepted our system constraints, as well as the terms of use of the tool. {% hint style="warning" %} **Text-To-3D Constraints are as follows:** * Describe a single object, creature or character in a noun phrase, adding only the adjectives you consider necessary, for example “blue cat” or “monkey with elephant head”. * Avoid over-complicating the prompt: if there are multiple ways of expressing your idea, use the simpler one. For example, “yellow submarine” is simpler than “submarine painted in yellow” or “submarine, yellow”. * Avoid non-English or very uncommon words, accents or other symbols when possible. Support for additional languages may come in the future. * Limit your text prompt to 100 characters. {% endhint %} When you're ready, hit Generate.

A completed text prompt form ready for generation

After pressing Generate, you will be taken to the generation progress screen, which shows you the status of your generation. This process can take up to six minutes.

A Text to 3D Model Generation in progress

Once your model has finished processing, you will be presented with the finished generation.

A generated model of a rabbit created using the prompt "white rabbit"

You can now decided to download the model as-is, generate another model, or animate the model using our Animate Anything tool. Your generated model will also be available to view in My World, under *generated models*, so you can access it any time! ## Finding Your Models in My World You can find all your successfully generated models in My World. To access this area, either select 'My World' from the dropdown menu when you hover over your profile image in the top right of the web application, or alternatively you can select 'My World' from the navigation on the left whilst on your personal Dashboard.

My World can be found within the dropdown menu that appears when hovering over your profile image in the top right of the navigation bar.

My World can also be found on your Dashboard, on the left-hand navigation above your account details

Once you navigate to My World, you can view your generated models by selecting the option 'Generated Models'. This will show you a truncated selection of your generated models. In order to view all models, be sure to select 'View All' on the right-hand side of the screen.

The generated models option will be shown with white text and a black background when selected.

When generated models is selected, you will be presented with all your models that you have created using Generate Anything.

{% hint style="info" %} You can tell which method was used to generate a model by referring to the icon on the top-left of the thumbnail. Generate Anything models that have been generated using Text-To-3D will have a ![](/files/Zpb8oEbS0j7VYNAMkg0s) symbol, whilst Generate Anything models that have been made using Image-To-3D will have a ![](/files/ky06YFcaTtm6UX3Jh6B8) symbol. {% endhint %} From the generated models screen, you can either select the model by clicking the thumbnail to be taken to the model viewer, where you can download your model, or you can press the 'Animate' button to be taken directly to the Animate Anything interface and start the process of animating the selected model.

## Downloading & Animating Your Model On the Model Viewer Screen you will be able to download your generated model. You can do this by pressing the button marked 'Download' {% hint style="info" %} Generated Models can be downloaded in the following formats: **FBX, GLB, GLTF, OBJ & USDZ** {% endhint %} If you wish to animate your model using our Animate Anything tool, all you need to do is press the button marked 'Animate Model' which will lead you through the Animate Anything interface. For further information on this process, [please refer to our documentation here.](/anything-world/quickstart/animate-anything-quickstart.md) --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://anything-world.gitbook.io/anything-world/quickstart/generate-anything-quickstart.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.