# 3D AI Models for Game Studios

> Compare 46 3D asset generation models on Layer — quality, capabilities, and per-generation pricing side by side. Generate production-ready assets for your game.

## 3D models (46)

- [Meshy Rigging](https://layer.ai/models/meshy-rigging) ([md](https://layer.ai/models/meshy-rigging.md)) by Meshy: Auto-rigs humanoid 3D models and optionally applies a preset animation, returning rigged GLB/FBX.
- [Meshy V6](https://layer.ai/models/meshy-v6) ([md](https://layer.ai/models/meshy-v6.md)) by Meshy: Meshy V6. High-fidelity 3D asset creation focusing on PBR, quad mesh, and face rigging.
- [Meshy V7](https://layer.ai/models/meshy-v7) ([md](https://layer.ai/models/meshy-v7.md)) by Meshy: Meshy V7. Text, image, and multi-image to 3D with PBR, quad mesh, and optional rigging.
- [Tripo v3.0](https://layer.ai/models/tripo3d-v3-0) ([md](https://layer.ai/models/tripo3d-v3-0.md)) by Tripo: Latest 3D model for production-quality assets with superior textures and clean geometry.
- [Hunyuan 3D 3.0](https://layer.ai/models/tencent-hunyuan3d-3-0) ([md](https://layer.ai/models/tencent-hunyuan3d-3-0.md)) by Tencent: Professional-grade 3D model optimized for high-quality, detailed assets with advanced features.
- [Rodin Gen-1.5](https://layer.ai/models/hyper3d-rodin) ([md](https://layer.ai/models/hyper3d-rodin.md)) by Hyper3D: Advanced 3D generation model creating high-quality, textured T-pose avatars from a single image.
- [Hi3D Parts](https://layer.ai/models/hitem3d-hi3d-parts) ([md](https://layer.ai/models/hitem3d-hi3d-parts.md)) by Hitem3D: Splits a 3D model into parts for editing, printing, or modular asset workflows.
- [Trellis 2](https://layer.ai/models/microsoft-trellis-2) ([md](https://layer.ai/models/microsoft-trellis-2.md)) (aka TRELLIS.2-4B) by Microsoft: Open-source, high-quality 3D model from Microsoft, leveraging a novel field-free sparse voxel structure.
- [Hunyuan 3D v3.1 Pro](https://layer.ai/models/tencent-hunyuan3d-v3-1-pro) ([md](https://layer.ai/models/tencent-hunyuan3d-v3-1-pro.md)) by Tencent: Latest Hunyuan 3D model with enhanced quality, multi-view input, and PBR material support.
- [Meshy Auto Split](https://layer.ai/models/meshy-auto-split) ([md](https://layer.ai/models/meshy-auto-split.md)) by Meshy: Splits a 3D model into separately printable parts, automatically or along parts you name.
- [Meshy UV Unwrap](https://layer.ai/models/meshy-uv-unwrap) ([md](https://layer.ai/models/meshy-uv-unwrap.md)) by Meshy: Lays out new UV coordinates on a 3D model, ready for texturing.
- [Meshy V6 Lite Retexture](https://layer.ai/models/meshy-v6-lite-retexture) ([md](https://layer.ai/models/meshy-v6-lite-retexture.md)) by Meshy: Retextures an uploaded 3D model from a text style or a reference image, with optional PBR maps.
- [Meshy V6 Retexture](https://layer.ai/models/meshy-v6-retexture) ([md](https://layer.ai/models/meshy-v6-retexture.md)) by Meshy: Retextures an uploaded 3D model with Meshy 6 from a text style or a reference image.
- [Meshy V7 Retexture](https://layer.ai/models/meshy-v7-retexture) ([md](https://layer.ai/models/meshy-v7-retexture.md)) by Meshy: Retextures an uploaded 3D model with Meshy 7 from a text style or a reference image.
- [Hunyuan 3D v3.1 Remesh](https://layer.ai/models/tencent-hunyuan3d-v3-1-remesh) ([md](https://layer.ai/models/tencent-hunyuan3d-v3-1-remesh.md)) by Tencent: Rebuilds a GLB or OBJ mesh with Hunyuan 3D smart topology at a low, medium, or high polygon density.
- [Tripo Smart Retopology v3](https://layer.ai/models/tripo3d-retopology-v3) ([md](https://layer.ai/models/tripo3d-retopology-v3.md)) by Tripo: Tripo's newest retopology: rebuilds a high-poly mesh as clean low-poly topology at a target face count, baking the original textures onto the result.
- [Tripo Smart UV](https://layer.ai/models/tripo3d-smart-uv) ([md](https://layer.ai/models/tripo3d-smart-uv.md)) by Tripo: Re-unwraps a mesh's UVs with AI-placed seams and packed islands, then remaps its existing textures, rig and animation onto the new layout. Geometry is kept as-is.
- [Tripo P2](https://layer.ai/models/tripo3d-p2) ([md](https://layer.ai/models/tripo3d-p2.md)) by Tripo: Tripo P2. Text or a single image to a textured 3D mesh with PBR, quad topology, and a 48 to 50,000 face budget.
- [Meshy V7.1](https://layer.ai/models/meshy-v7.1) ([md](https://layer.ai/models/meshy-v7.1.md)) by Meshy: Meshy V7.1. Text, image, and multi-image to 3D with PBR, quad mesh, and optional rigging.
- [Hunyuan 3D UV Unwrap](https://layer.ai/models/tencent-hunyuan3d-uv-unwrap) ([md](https://layer.ai/models/tencent-hunyuan3d-uv-unwrap.md)) by Tencent: Rebuilds an existing mesh's UV layout so it is ready to texture. Geometry is kept as-is and the result is untextured.
- [Hunyuan 3D v3.1 Part](https://layer.ai/models/tencent-hunyuan3d-v3-1-part) ([md](https://layer.ai/models/tencent-hunyuan3d-v3-1-part.md)) by Tencent: Splits a 3D FBX model into separate parts with Hunyuan 3D v3.1.
- [Hi3D v3.0](https://layer.ai/models/hitem3d-hi3d-v3) ([md](https://layer.ai/models/hitem3d-hi3d-v3.md)) by Hitem3D: Hi3D v3.0 image-to-3D at 2048³ with PBR materials and face counts up to 5M.
- [Uthana Animate](https://layer.ai/models/uthana-animate) ([md](https://layer.ai/models/uthana-animate.md)) by Uthana: Retargets a Uthana motion — trained, or generated from a prompt or video — onto your character.
- [Uthana Auto-Rig Character](https://layer.ai/models/uthana-auto-rig-character) ([md](https://layer.ai/models/uthana-auto-rig-character.md)) by Uthana: Uthana's automatic character rigging model, preparing an uploaded mesh with a skeleton.
- [Meshy T2](https://layer.ai/models/meshy-t2) ([md](https://layer.ai/models/meshy-t2.md)) by Meshy: Meshy T2. Text or single image to a compact, cleanly built multi-part mesh with textures, PBR, and optional rigging.
- [Rodin Bang](https://layer.ai/models/hyper3d-bang) ([md](https://layer.ai/models/hyper3d-bang.md)) by Hyper3D: Segments a 3D mesh into parts with Hyper3D Bang!
- [Rodin Gen-2](https://layer.ai/models/hyper3d-rodin-gen-2) ([md](https://layer.ai/models/hyper3d-rodin-gen-2.md)) by Hyper3D: Hyper3D Rodin Gen-2 delivers sharper geometry and cleaner textures from a single image or prompt.
- [Rodin Gen-2.5](https://layer.ai/models/hyper3d-rodin-gen-2-5) ([md](https://layer.ai/models/hyper3d-rodin-gen-2-5.md)) by Hyper3D: Hyper3D Rodin Gen-2.5 pushes structural detail and surface quality further, with selectable quality.
- [Hi3D](https://layer.ai/models/hitem3d-hi3d) ([md](https://layer.ai/models/hitem3d-hi3d.md)) by Hitem3D: Hi3D image-to-3D with multi-view support, PBR materials, and configurable face counts.
- [Hi3D Texture](https://layer.ai/models/hitem3d-hi3d-texture) ([md](https://layer.ai/models/hitem3d-hi3d-texture.md)) by Hitem3D: Textures an existing Hi3D-compatible geometry mesh from a reference image.
- [Hunyuan 3D v3.1 Fast](https://layer.ai/models/tencent-hunyuan3d-v3-1-fast) ([md](https://layer.ai/models/tencent-hunyuan3d-v3-1-fast.md)) by Tencent: Fast Hunyuan 3D model for rapid 3D generation with PBR material support.
- [Tripo P1](https://layer.ai/models/tripo3d-p1) ([md](https://layer.ai/models/tripo3d-p1.md)) by Tripo: Game-ready Tripo model: a single image to clean low-poly 3D with PBR textures.
- [Tripo v3.1](https://layer.ai/models/tripo3d-v3-1) ([md](https://layer.ai/models/tripo3d-v3-1.md)) by Tripo: High-detail Tripo model for production assets with dense geometry and refined textures.
- [Bytedance Seed 3D](https://layer.ai/models/bytedance-seed3d) ([md](https://layer.ai/models/bytedance-seed3d.md)) by ByteDance: Powerful 3D model focusing on high-quality objects from a single image. Adept at geometry & texture.
- [SAM 3D Align 1.0](https://layer.ai/models/meta-sam-3-3d-align) ([md](https://layer.ai/models/meta-sam-3-3d-align.md)) by Meta: Meta SAM 3D Align places body and object meshes into a spatially coherent scene from one image.
- [SAM 3D Body 1.0](https://layer.ai/models/meta-sam-3-3d-body) ([md](https://layer.ai/models/meta-sam-3-3d-body.md)) by Meta: Meta SAM 3D Body reconstructs accurate human body shape and pose from a single image.
- [SAM 3D Objects 1.0](https://layer.ai/models/meta-sam-3-3d-objects) ([md](https://layer.ai/models/meta-sam-3-3d-objects.md)) by Meta: Meta SAM 3D Objects reconstructs textured 3D geometry from a single real-world image.
- [Tripo Animate](https://layer.ai/models/tripo3d-animate) ([md](https://layer.ai/models/tripo3d-animate.md)) by Tripo: Rigs a 3D mesh and applies a preset animation from the animation library, returning an animated GLB.
- [Tripo Remesh](https://layer.ai/models/tripo3d-remesh) ([md](https://layer.ai/models/tripo3d-remesh.md)) by Tripo: Rebuilds a high-poly 3D mesh as a low-poly model at a target face count, baking the original textures onto the result.
- [Tripo Retexture](https://layer.ai/models/tripo3d-retexture) ([md](https://layer.ai/models/tripo3d-retexture.md)) by Tripo: Retextures an existing mesh from a text prompt or reference image, keeping geometry intact.
- [Tripo Rig](https://layer.ai/models/tripo3d-rig) ([md](https://layer.ai/models/tripo3d-rig.md)) by Tripo: Auto-rigs a 3D mesh, returning a rigged GLB. Humanoid meshes come back with a Mixamo-named skeleton for direct use in game engines; other body plans use Tripo's own naming, which Mixamo cannot describe.
- [Tripo Segment](https://layer.ai/models/tripo3d-segment) ([md](https://layer.ai/models/tripo3d-segment.md)) by Tripo: Segments a 3D mesh into parts, returning a part-grouped model for further editing.
- [Hunyuan 3D v2 Mini](https://layer.ai/models/tencent-hunyuan3d-v2-mini) ([md](https://layer.ai/models/tencent-hunyuan3d-v2-mini.md)) (aka Hunyuan3D-2mini) by Tencent: Lightweight, efficient 3D version optimized for less powerful hardware. Delivers good quality assets.
- [Tripo v2.5](https://layer.ai/models/tripo3d-v2-5) ([md](https://layer.ai/models/tripo3d-v2-5.md)) by Tripo: Incremental 3D update. Refined performance, improved mesh topology, and texture fidelity.
- [Hunyuan 3D v2](https://layer.ai/models/tencent-hunyuan3d-v2) ([md](https://layer.ai/models/tencent-hunyuan3d-v2.md)) (aka Hunyuan3D 2.0) by Tencent: Powerful, open-source 3D model producing high-res, textured 3D objects from text or image inputs.
- [Trellis](https://layer.ai/models/microsoft-trellis) ([md](https://layer.ai/models/microsoft-trellis.md)) by Microsoft: Open-source 3D model creating high-quality objects with realistic materials and geometry from text.

## Frequently Asked Questions

### Which is the best AI 3D model?

Meshy Rigging by Meshy is the most-used 3D model on Layer. The Artificial Analysis Arena does not score 3D models, so this catalogue ranks them by use rather than by Elo.

### How are 3D models ranked on this page?

3D models are ranked by their Artificial Analysis Arena Elo: people compare two generations made from the same prompt without being told which model made which, and a model preferred more often scores higher. An Elo is a standing inside one modality, so models are only ever ranked against others of their own kind. A model with no arena score yet sits below the scored ones, ordered by how much it is used on Layer and how recently it was released.

### How many 3D models does Layer have?

Layer gives you 46 3D asset generation models from 9 providers — Tripo, Meshy and Tencent among them — in one workspace, on one balance, behind one API.

### Which 3D models can I train on my own artwork?

Uthana Animate can be fine-tuned on a studio's own art. Upload a reference set, train a style, and every generation from it carries your look rather than the model's default one.

### How much do 3D models cost?

Nothing is charged per seat. Generating on Layer spends Creative Units, each model has its own rate, and you pay only for what you generate — every model page states its current price, and the pricing page lists them all.

---

[View original](https://layer.ai/models/3d) | [All models](https://layer.ai/models)