Tripo Releases Segmentation V2: Smarter, Cleaner Part Separation for AI-Generated 3D Models

Tripo, one of the leading AI-powered 3D generation platforms, has rolled out Segmentation V2 in its May update. The new version significantly improves how AI-generated 3D objects are broken down into meaningful semantic parts — and it does so based purely on the model’s geometry.

This upgrade delivers noticeably cleaner boundaries between segments, making the resulting models far more practical for editing, texturing, 3D printing, game asset pipelines, and further sculpting workflows.

How Segmentation V2 Works

Unlike earlier versions that could produce fuzzy or inconsistent part divisions, Segmentation V2 analyzes the underlying geometry to identify logical, semantic components (such as limbs, torso, accessories, or mechanical parts).

Users can now control the level of detail with three precision modes:

Simple — Broad splits (roughly 3–6 major parts);
Balanced — Medium granularity (6–15 parts);
Detailed — High granularity (15+ parts) — ideal for complex assets needing fine separation for games or manufacturing.

The result is much sharper, more professional-looking segmentation with minimal manual cleanup required.

Real-World Performance and Limitations

Tripo Releases Segmentation V2: Smarter, Cleaner Part Separation for AI-Generated 3D Models Early user tests show strong results on most organic and mechanical shapes. However, fine surface details like text on clothing (e.g., logos or lettering on a t-shirt) can still struggle at lower input resolutions. One tester noted that uploading a 720p image didn’t produce clean text segmentation, but expects significantly better results with higher-resolution source images.

This highlights that while the geometry-based approach is a big leap forward, input quality still matters for intricate surface features.

Other Major May Updates

Alongside Segmentation V2, Tripo introduced several powerful new capabilities:

True 8K Textures — The platform can now generate photorealistic 8K textures in under two minutes. These work on any model (Tripo-generated or imported from elsewhere) and are export-ready for major 3D applications. Every user receives one free 8K texture generation as part of the update.
Tripo DCC Bridge for ZBrush — The first native bridge from any AI 3D generation tool directly into ZBrush. Users can generate a model in Tripo and immediately begin sculpting in ZBrush without exporting files or dealing with format conversions. This is a major workflow win for professional 3D artists who rely on ZBrush for final detailing.
Quick Cap (AI Completion) — Automatically seals open edges after segmentation, speeding up the transition from split model to a clean, watertight mesh.

Why This Matters

Tripo Releases Segmentation V2: Smarter, Cleaner Part Separation for AI-Generated 3D Models Segmentation has long been one of the biggest pain points in AI-generated 3D content. Raw outputs are often monolithic and difficult to work with in traditional pipelines. By delivering geometry-aware, controllable part separation with clean boundaries, Tripo is making its models significantly more production-ready.

Combined with 8K texturing and direct ZBrush integration, the platform is evolving from a “generate-and-export” tool into a more complete end-to-end creation environment.

Availability

As part of the May update rollout, Tripo is offering one free Segmentation + 8K texture generation to all users. The company is also running its biggest sale to date (up to 75% off) through the end of June, along with new subscription tiers including a “Max” plan for heavy individual users and a new Team plan for collaborative workflows.

Tripo continues to push the boundaries of accessible, high-quality AI 3D generation. With Segmentation V2, 8K texturing, and native DCC integrations, it’s becoming increasingly viable for serious creative and professional use cases.

Try Tripo and get a QUA bonus: https://quasa.io/projects/tripo-ai