Generating 3D models from text or images
Video-generation AI company Runway said through its official account on X on Nov. 12 (11:15:57 p.m. Nov. 12 Japan time) that its service can now handle 3D. 3D models are generated with Tripo's "Tripo H3.1" model. According to the company, entering text or images produces detailed 3D models. Users can also upload and use 3D models they have prepared themselves. The post said "Try it now in Runway," indicating the feature is already available.
3D is now supported in Runway. Generate detailed 3D models from text or images with Tripo H3.1, or upload your own. Choose any angle and use it as a reference for your next image or video. Export your model to use in other 3D tools. Try it now in Runway.
Any angle as a reference for images and video
According to the post, users can choose a view of a 3D model from any angle and use it as a reference for the next image or video they create. A reference is an example used to match the look of the generated image or video. Models can also be exported for use in other 3D tools. The post did not list the file formats available for export or the formats that can be uploaded.
10 to 50 credits per generation
Runway's pricing and model list page describes the use of Tripo H3.1 as "Generating a 3D model from text or images" and lists two modes, "Text to 3D" and "Image to 3D." One generation uses 10 credits at the "Standard" setting and 30 at the more detailed "Detailed" setting, rising when surface texture is added, up to a maximum of 50.
Neither the post nor the pricing page states which plans are eligible. As of 12:21 a.m. Nov. 13 Japan time, Runway's changelog did not mention 3D support.
| Setting | Credits used |
|---|---|
| Standard | 10 |
| Standard + texture (standard) | 20 |
| Standard + texture (detailed) | 30 |
| Detailed | 30 |
| Detailed + texture (standard) | 40 |
| Detailed + texture (detailed) | 50 |
