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

# Generate new views with 3D models

> Convert a concept image to a 3D model, capture new angles in Rhino, then re-render them to match the original photo

Image-to-3D models give you real geometry you can orbit, but the textures are rough. Use the 3D model only for the camera and shape, then re-render each view to match the original image.

![A cardboard concept, four rough views of its 3D model, and the four views re-rendered as photographs](https://mintcdn.com/runchat/ExzxwftWKPCOoCF3/images/examples/new-views-with-3d-models.webp?fit=max\&auto=format\&n=ExzxwftWKPCOoCF3\&q=85\&s=2ab3dbe8145031e90e390b6e2b68712a)

> [card] [Open the example workflow](https://runchat.com/editor?id=15191947-babe-4d2d-84e9-68b684d08671)
>
> Make a copy of this workflow in your account to try it with your own images.

## How it works

1. **Generated concept (Flux 2 Klein)** makes the image.
2. **Image to 3D (Tripo H3.1)** converts it into a textured 3D model.
3. **(optional) Rhino: Import the 3D model into Rhino** brings the model in, and **Four new angles** captures four views as a 2x2 sheet.
4. **Re-render the four views to match the photo (Flux 3)** turns the rough views into photographs of the real model, and **Split into four views** separates them.

## Tips

* Describe the original's materials and setting in the re-render prompt. The 3D textures are too blurry to carry them.
* See [Generate 3D models from images](https://docs.runchat.com/examples/generating-3d-models.md) for choosing an image-to-3D model.

## Use your own model

The image input holds a capture from our demo Rhino model, so the workflow runs as it is. To use your own model, open Rhino with the [Runchat plugin](https://docs.runchat.com/plugins/rhino.md) signed in, set up your view, and run the node marked **(optional) Rhino**. Then drag its screenshot into the image input. The scripts use only the standard display modes and match layers or materials by the keywords in their inputs, so they work on any model.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.