Production-Scale 3D Model Generation API

Add 3D generation to your backend with the Hyper3D API. Powered by Rodin, this 3D model generation API supports text-to-3D, image-to-3D, and texturing, with every task tracked from submission to download.

generate-model.ts
curl https://api.hyper3d.com/api/v2/rodin\-H"Authorization: Bearer $RODIN_API_KEY"\-F"tier=Gen-2" \-F"prompt=stylized sci-fi crate" \-F"geometry_file_format=glb"

Get Started with the Rodin API

Everything you need to go from API key to production, from your first request to texturing, plug-ins, and usage tracking.

Build with the Rodin API

Use one AI 3D model API for text-to-3D, image-to-3D, and texturing existing models. Choose the workflow you need, copy the sample request, and start building.

Examples


                              curl     --fail-with-body   \        --request   POST   'https://api.hyper3d.com/api/v2/rodin'   \        --header     "Authorization: Bearer ${RODIN_API_KEY}"   \        --form     'prompt=A small wooden treasure chest with brass trim'   \        --form     'tier=Gen-2.5-Medium'   \        --form     'mesh_mode=Raw'   \        --form     'quality=medium'   \        --form     'geometry_file_format=glb'   \        --form     'material=PBR'      
                          

An API Plan for Every Stage

Compare rate limits, concurrency, and access across plans.

An API Plan for Every Stage
Capability Business Enterprise popular
Scale & Performance Requests per minute 120–240
Concurrent tasks 5
Task priority Standard
Uptime SLA Not available
Data & Security Data retention 7 days
Security certification Not available
Customization & Support Usage and request logs Included
Custom endpoints and LoRA fine-tuning Not available
On-premises deployment Not available
Dedicated account support Not available

AI 3D generation, built into your product.

Integrate Rodin into your app, pipeline, or tool. Get your API key and start building.

Get an API Key

Pricing

Usage-based pricing. Pay only for the 3D generation you use.

Rodin Gen-2.5

From 0.5 credit / generation
  • Custom Polycount Up to 2M
  • Up to 12K Textures with HighPack
  • Quad or Raw Topology
  • GLB, FBX, OBJ, USDZ, STL Export
  • Text, single-image, or up to 5-image input

Rodin Gen-2

From 0.5 credit / generation
  • Production-Ready 3D Assets
  • Flexible Mesh Quality
  • PBR Materials & Realistic Textures
  • Text & Image-to-3D Generation
  • Industry Standard Export Formats

HighPack

From + 1 credit / generation
  • Up to 12K Textures
  • High-Poly 3D Meshes
  • Enhanced Surface Details
  • Premium Rendering Quality
  • Professional Production Assets

Rodin Task Bang

From 0.5 credit / generation
  • Split any Rodin model into parts
  • Choose from 1 to 12 split levels
  • Guide the split with a text instruction
  • GLB, FBX, OBJ, USDZ, or STL Export

Custom Model Bang

From 0.5 credit / generation
  • Split your own OBJ, GLB, FBX, STL, or USD model
  • Choose from 1 to 12 split levels
  • Guide the split with a text instruction
  • GLB, FBX, OBJ, USDZ, or STL Export

Texture Generation

From 0.5 credit / generation

Input

  • 3D Model
  • Reference Image

Output

  • PBR, Shaded, or Both
  • Optional Textured Model
  • Up to 12K on Extreme-High

Options

  • Texture Resolution: 1K/2K/4K
  • Material Type: PBR, Stylized
  • Quality: Basic, High

3D Generation API: Frequently Asked Questions

Can I Generate Models from Both Text and Images?

Yes. With the text to 3D API, send a prompt of up to 1,024 characters. With the image to 3D API, upload 1–5 images of the same object and optionally add a prompt to guide the result.

Which Plans Include API Access?

The Hyper3D API is available on Business and Enterprise plans. Business includes standard API access and support, while Enterprise adds custom limits, a 99.9% uptime SLA, and dedicated support. Once you’re on an eligible plan, create your key in the API dashboard.

Does the API Return a Model Immediately?

No. The Rodin API runs every request as an async job. Submitting a task returns a task ID and a subscription key right away. Poll the status endpoint with the subscription key until every job shows Done, then download your files with the task ID.

What Happens If a Task Fails?

A failed task returns a Failed status and no output files. Requests may also return errors such as 400 for invalid input, 401 for an invalid API key, or 429 when rate or concurrency limits are reached. The Errors page lists every error code and how to fix it.

How Is API Usage Charged?

Our 3D model API pricing is usage-based, with generation credits deducted from your plan. Status checks, downloads, and balance checks do not consume credits, and usage can be reviewed by task.

Which Formats Can I Download?

Generated models can be exported as GLB, FBX, OBJ, USDZ, or STL. Use GLB or USDZ for web and AR, FBX or OBJ for DCC tools and game engines, and STL for 3D printing.

What Are the Rate Limits and Retention Rules?

Business plans support 120–240 requests per minute, depending on the tier. Enterprise limits are customized. Uploaded inputs and generated files are retained for 7 days.

Should I Use the API or Hyper3D MCP?

Use the 3D model generation API when your app, script, or backend needs direct control over requests, settings, and files. Use Hyper3D MCP when you want an AI agent such as Codex, Claude Code, or Kimi Code to generate 3D models directly from chat. Both use the same underlying Rodin models.

Build 3D Generation into Your Product

Get your API key, send your first request to the 3D model generation API, and ship Rodin-powered 3D into your app, tool, or pipeline.