Working effectively with AI tools requires a hybrid approach—currently, these tools don't perfectly handle all asset types. Here are some guidelines to help you maximize your results:
Setting Expectations
For most use cases, a 3D modeler or artist will need to perform final adjustments on Cube’s outputs, such as retopology and retexturing. The main objective of using Cube is to streamline your workflow and save significant time. For practical examples of Artist x Cube workflows, check out our 3D Workflow Demos on the CSM YouTube channel.
Complex Assets
To fully leverage Cube's capabilities, break complex assets into smaller components, generate them individually, and then reassemble. If you're getting subpar results, follow our detailed guide: Generating Complex Meshes.
Batch Upload
If testing Cube on multiple assets, we highly recommend batch uploading to reduce waiting times. Refer to our guide here: How to Generate Multiple Assets at Once Guide.
Text & Logos
AI tools currently have difficulties accurately rendering text and logos. When uploading meshes containing these elements, expect significant visual discrepancies. Fortunately, there is a quick fix—see our Fixing Logos and Text Elements Guide.
Feedback
We welcome your feedback, questions, or suggestions to continually enhance our tools and resources. For immediate assistance or to set up a private Slack channel for faster communication, contact us at hello@csm.ai.
Example Use Cases:
Gaming: Effective for props, environments, pitch materials, and live-ops content; less effective for complex characters.
3D Products: Strong performance in staging assets, furniture, products, and clothing across various categories.
