Loading model

About

A 2 m stone bench run 540 mm high: a marble seat slab on two chamfered travertine piers with a red band under the front edge. Runs butt end to end along a wall, a colonnade or a changing room. Metre units, +Y up, the bather's side toward +Z, resting on y=0. Static geometry; no rig, collision or navigation data.

Use with an AI agent

Copy this prompt into your AI chat, then describe what you want to build with the asset.

Use the “Stone Bench Run (Roman Bathhouse and Palaestra)” asset in my project.

Model: https://cdn.3dassets.dev/assets/37891/v1/model.glb
Asset page: https://3dassets.dev/assets/roman-bathhouse-and-palaestra-stone-bench-run-565d045c
Description: A 2 m stone bench run 540 mm high: a marble seat slab on two chamfered travertine piers with a red band under the front edge. Runs butt end to end along a wall, a colonnade or a changing room.

- Dimensions (X × Y × Z): 2 × 0.54 × 0.56 metres
- Triangles: 204
- Animations: none
- Licence: CC0 1.0 Universal (https://creativecommons.org/publicdomain/zero/1.0/)
- Loading: Three.js GLTFLoader; no additional decoders required.

Also available as JSON, Markdown, or via the MCP server.

Import manually

Choose your framework, then copy the code into your project to load the asset.

import { GLTFLoader } from 'three/addons/loaders/GLTFLoader.js'

const loader = new GLTFLoader()
loader.load('https://cdn.3dassets.dev/assets/37891/v1/model.glb', (gltf) => {
  scene.add(gltf.scene)
})

Common questions

How do I use this model with my AI development tools?

Press "Copy for AI agent" above and paste the prompt into Claude Code, Cursor, Codex, GitHub Copilot, ChatGPT or any other coding assistant. The prompt carries the direct CDN URL, the real-world size, the licence and the loading notes, so the assistant can fetch the model and place it in your project without you writing the loader yourself. For a permanent connection, add the 3DAssets.dev MCP server to your tool with one command (for Claude Code: claude mcp add --transport http 3dassets https://3dassets.dev/mcp) and the assistant can then search the whole catalogue, fetch files and read the usage snippets on its own.

MCP server setup Guide for AI agents

How do I get started?

Three ways, from quickest to most automatic. Download the .glb with the button and drop it into your engine's asset folder, where it imports like any glTF file. Or skip the download: copy the three.js, React Three Fiber or model-viewer snippet under "Import manually", which loads the model straight from the CDN. Or connect your AI agent to the MCP server and ask for it by name, and it will find, fetch and place the model for you.

MCP server setup API reference

Which game engines and tools can I use this model with?

Anything that reads glTF, which today means all of the major ones: three.js and React Three Fiber on the web, Godot 4, Unity, Unreal Engine 5, Blender and the model-viewer web component. The model is a self-contained binary glTF (.glb) with textures embedded, so there is one file to import and no plugin, converter or decoder to install first.

How do I import this model into Unity, Godot, Unreal or Blender?

Unity: install the glTFast package (com.unity.cloud.gltfast) from the Package Manager, drop the .glb into Assets and drag the prefab in, or load it at runtime from the CDN URL. Godot 4: copy the file into the project folder and the editor imports it as a scene you can instantiate. Unreal Engine 5: File > Import with the built-in glTF importer, which converts the axes for you. Blender: File > Import > glTF 2.0. Units are metres with +Y up in the file, and every importer above handles the conversion to its own axes automatically.

Can I use this model in a commercial game, and do I need to credit anyone?

Yes to commercial use and no to credit. Every model on 3DAssets.dev is released under CC0 1.0 Universal, which places it in the public domain: use it in a paid game, an app, a film or a client project, modify it, redistribute it, and put nothing in the credits. A link back is welcome but it is entirely your choice.

CC0 licence text

Can I load it from the CDN instead of downloading it?

Yes. The .glb file has a permanent URL on cdn.3dassets.dev that is CORS-enabled and never changes content, so you can pass it straight to GLTFLoader, glTFast or model-viewer in production and cache it as hard as you like. Downloading is only needed for engines that import from disk.

Do I need Draco, Meshopt or KTX2 to load it?

No. The geometry is quantized with KHR_mesh_quantization and the textures are re-encoded to WebP, both of which three.js, Godot, Unity and Blender read natively, so a plain GLTFLoader is enough. There is nothing extra to configure and no decoder to ship with your bundle.

Stone Bench Run (Roman Bathhouse and Palaestra)

A 2 m stone bench run 540 mm high: a marble seat slab on two chamfered travertine piers with a red band under the front edge. Runs butt end to end along a wall, a colonnade or a changing room.

Download asset · 11.1 KB

More in Props & Furniture

Mesh Stillage (Car Factory Production Line)

Free 3D model of a stackable mesh stillage holding loose parts, a factory logistics container for a car factory game asset kit.

@3dassets4.4k tris · 10 ↓