Trellis 2
Also known as TRELLIS.2-4B
Trellis 2 is a state-of-the-art large 3D generative model designed for high-fidelity image-to-3D generation. It leverages a novel field-free sparse voxel structure termed O-Voxel and a large-scale flow-matching transformer (4 Billion parameters). Unlike previous methods that rely on iso-surface fields (e.g., SDF, Flexicubes) which struggle with open surfaces or non-manifold geometry, TRELLIS can reconstruct and generate arbitrary 3D assets with complex topologies, sharp features, and full Physical-Based Rendering (PBR) materials—including transparency/translucency.
See more
Trellis 2 is a state-of-the-art large 3D generative model designed for high-fidelity image-to-3D generation. It leverages a novel field-free sparse voxel structure termed O-Voxel and a large-scale flow-matching transformer (4 Billion parameters). Unlike previous methods that rely on iso-surface fields (e.g., SDF, Flexicubes) which struggle with open surfaces or non-manifold geometry, TRELLIS can reconstruct and generate arbitrary 3D assets with complex topologies, sharp features, and full Physical-Based Rendering (PBR) materials—including transparency/translucency.
Connect Claude, Cursor, VS Code, or Codex to Layer’s MCP server and Trellis 2 becomes a tool your agent can reach for — it picks the parameters, runs the generation, and shows you the result in the conversation.
https://mcp.app.layer.ai/mcpAsk your agent
“Use Trellis 2 to model a wooden treasure chest with iron bands, game-ready.”
execute_forge({
"base_model_id": "trellis-2",
"prompt": "A wooden treasure chest with iron bands, game-ready",
"batch_size": 1
})Authenticate with a personal access token, post the model’s own form, and poll the inference until it completes. Every field below is accepted by this endpoint.
# Start the generation — returns an inference id straight away.
curl -X POST https://api.app.layer.ai/api/v2/workspaces/$WORKSPACE_ID/base-models/microsoft-trellis-2/inferences \
-H "Authorization: Bearer $LAYER_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"guidance_file_init_image": [
{
"url": "https://example.com/input.png"
}
]
}'
# Poll until status is "complete" — the response then carries the asset URLs.
curl https://api.app.layer.ai/api/v2/workspaces/$WORKSPACE_ID/inferences/$INFERENCE_ID \
-H "Authorization: Bearer $LAYER_TOKEN"- seed integer
- The starting point for the model's randomness. The same seed and settings give the same result.
- guidance_file_init_image file[]required
- The image the 3D model is built from
Why use Trellis 2 for 3D?
High-quality 3D generation
Trellis 2 produces clean geometry, accurate proportions, and detailed surfaces suitable for game-ready assets and prototyping.
Engine-ready output
Exports in standard formats compatible with Unity, Unreal Engine, and Blender. PBR materials and clean topology where supported.
Rapid prototyping
Go from concept to 3D asset in minutes. Ideal for early-stage exploration, shop items, and LiveOps content.
Rapid 3D asset prototyping for game development
Go from concept image to 3D model in minutes. Ideal for evaluating designs, testing in-engine, and building asset libraries.
Shop items and cosmetic content for LiveOps
Generate 3D items, weapons, and cosmetics for live service updates. Keep your content pipeline flowing without bottlenecks.
Marketing renders and promotional 3D content
Create polished 3D renders for store listings, social media, and ad creatives without heavy production overhead.
FAQ
How much does Trellis 2 cost? Is it free?+
How does Layer's pricing work?+
There are no seat fees, feature gates, or plans on Layer. Instead, our platform uses a consumption based system with Creative Units (CUs) with a flexible monthly subscription. Every generation on Layer (image, video, 3D, or audio) consumes a Creative Unit, and you only pay for what you create.
Other 3d models on Layer
Trellis
Microsoft
Open-source 3D model creating high-quality objects with realistic materials and geometry from text.
Hunyuan 3D v3.1 Pro
Tencent
Latest Hunyuan 3D model with enhanced quality, multi-view input, and PBR material support.
Hi3D Parts
Hitem3D
Splits a 3D model into parts for editing, printing, or modular asset workflows.
Meshy Auto Split
Meshy
Splits a 3D model into separately printable parts, automatically or along parts you name.
Rodin Gen-1.5
Hyper3D
Advanced 3D generation model creating high-quality, textured T-pose avatars from a single image.
Meshy UV Unwrap
Meshy
Lays out new UV coordinates on a 3D model, ready for texturing.