Create a 3D model

Describe one object and get back a self-contained glTF binary with textures, a matching picture of the object, and a provenance file that records exactly how it was made. Choose economy for a compact model in minutes or high for full detail. Geometry is unitless: no physical size, rigging, watertightness or print readiness is inferred from a description.

Input

A description of up to 4,000 characters; no file

Outputs

model.glb, model.web.glb, reference.png and generation.json

Quality

economy: about 40,000 faces, 1024 × 1024 PBR textures · high: 2048 × 2048 textures, about one million faces

Duration

Asynchronous; one-hour deadline including queueing and execution

Send the options as JSON. The response is the new job; poll it until its status is final, then download the files in result.files.

# idempotency_key is required: repeating the same request returns the same job
curl https://api.mcpbytes.com/v1/tools/create_3d_model/jobs \
  -H "Authorization: Bearer $MCPBYTES_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "options": {
      "description": "a brass bankers desk lamp with a green glass shade",
      "quality": "high",
      "seed": 3
    },
    "idempotency_key": "model-example-001"
  }'

{
  "tool": "create_3d_model",
  "status": "queued",
  "input_kind": "parameters",
  "options": {
    "description": "a brass bankers desk lamp with a green glass shade",
    "quality": "high",
    "texture_resolution": 2048,
    "seed": 3
  },
  "workflow": {
    "deadline": "2026-09-22T09:19:17.623Z",
    "cancel_pending": false,
    "refund_pending": false
  },
  "capabilities": {
    "can_cancel": true
  },
  "result": null
}

# JOB_ID is the id from the create response; poll until status is final
curl "https://api.mcpbytes.com/v1/jobs/$JOB_ID" \
  -H "Authorization: Bearer $MCPBYTES_API_KEY"

# Once status is succeeded, download the model from result.files[].url
curl -fsSLo model.glb "$MODEL_URL"

{
  "tool": "create_3d_model",
  "status": "succeeded",
  "credits_charged": 120,
  "compute_ms": 157059,
  "result": {
    "files": [
      {
        "name": "reference.png",
        "bytes": 500122,
        "content_type": "image/png",
        "url": "https://api.mcpbytes.com/blob/dl/…"
      },
      {
        "name": "model.glb",
        "bytes": 43029520,
        "content_type": "model/gltf-binary",
        "url": "https://api.mcpbytes.com/blob/dl/…"
      },
      {
        "name": "generation.json",
        "bytes": 2796,
        "content_type": "application/json",
        "url": "https://api.mcpbytes.com/blob/dl/…"
      }
    ]
  }
}

Jobs of this tool take minutes. While one runs, workflow.stage says what is happening and workflow.deadline when it must finish. Cancel it with POST /v1/jobs/{id}/cancel while capabilities.can_cancel is true.

More on statuses, downloads and retention in Jobs & files.

No file. Send the options in a JSON body, {"options": {…}}, or as arguments of the MCP tool. Keep private text such as prompts out of URLs.

Put options in the options object of the JSON body, or pass them as arguments of the MCP tool. Leave an option out to use its default.

A succeeded job lists its files in result.files, each with a download url.

Download links last up to 24 hours, capped by output retention; read the job again for fresh links. The model is a best-effort interpretation of the description: sides the picture does not show are invented, dimensions are not physical, and materials use opaque rendering.

• Describe one object with its materials and colours. Avoid scenes and text.
• Send the description in the JSON body or as MCP arguments, never in a URL.
• Hidden sides are invented, and geometry is unitless: no physical size, rigging, watertightness or print readiness is inferred.
• A job takes minutes: keep its id and poll get_job. Repeating the request with the same idempotency key returns the same job instead of a new charge.

Your agent calls create_3d_model. Connect it to https://api.mcpbytes.com/mcp/3d for this tool and the rest of its family (Create a 3D model from an image, Optimize a 3D model, Simplify a 3D model, Split a 3D model into parts, Inspect a 3D model), or to https://api.mcpbytes.com/mcp for every tool.

claude mcp add --transport http mcpbytes \
  https://api.mcpbytes.com/mcp/3d

codex mcp add mcpbytes --url https://api.mcpbytes.com/mcp/3d \
  --bearer-token-env-var MCPBYTES_API_KEY

{
  "mcpServers": {
    "mcpbytes": {
      "url": "https://api.mcpbytes.com/mcp/3d",
      "headers": {
        "Authorization": "Bearer ${env:MCPBYTES_API_KEY}"
      }
    }
  }
}

Arguments

The tool's options are flat arguments. Besides them:

What the agent sees

Setup for every client, and how agents upload files and read results: Connect over MCP.

The MCPBytes skill teaches agents this tool, in references/create_3d_model.md. Agents without MCP run it through the REST API with the skill's script, which reads your key from MCPBYTES_API_KEY:

printf '{"description": "a wooden rocking horse",
  "quality": "economy"}' > options.json
python scripts/mcpbytes.py run create_3d_model \
  --options-json options.json --out model/
python scripts/mcpbytes.py job j_...

From 50 credits. The selected option's price is charged before paid processing starts, and a canceled job is refunded. A job that fails costs nothing: its charge is refunded.

One credit balance pays for every tool, and a new account starts with free credits. See Credits & prices or buy credits.

GET /v1/me returns your limits and current usage; Limits & errors explains what happens when one is reached.

A failed job comes back with error.code, error.message and often error.hint. Codes are stable. Inputs that fail with a validation error will fail again unchanged.

Requests can also fail before a job starts, with an HTTP status: see HTTP errors.

• Optimize a 3D modelRebuild model.glb at the triangle count you need: pass it as job_file.optimize_3d_model
• Simplify a 3D modelReduce the triangle count and keep the textures: download model.glb and upload it.simplify_3d_model
• Split a 3D model into partsSplit the model into parts: download model.glb and upload it.split_3d_model
• Inspect a 3D modelCheck counts, bounds and topology: download model.glb and upload it.inspect_3d_model