The character pipelineas an API

Generate a model, rig it, retarget motion onto it, pull from a CC0 asset library. One job model across every step, driven from your own code or from an AI agent.

Quickstart

API reference

Get an API key

One job model

Rigging and generation return the same job object with the same statuses and one polling endpoint. Chain them by id, without moving files through your own storage.

How jobs work

Rigging

POST a mesh, get back a rigged, skinned GLB or FBX with Mixamo-compatible bones. Bipeds and creatures, typically 15 to 30 seconds.

Rig a model

Retargeting

Bake motion onto a character server-side, or download the clips and retarget free in your own engine. 260 CC0 animations either way.

Animate a character

Asset library

Nearly 7,000 CC0 props, searchable by category and pack, served from our CDN with no attribution required.

Browse assets

Agent-ready

An MCP server exposes rigging, assets, and animation to Claude and other AI agents, with OAuth and no browser in the loop.

Connect an agent

What this is

Cinevva's platform covers the unglamorous middle of 3D character work: taking a static mesh and making it a thing your engine can actually play animations on. It is the same pipeline behind the Cinevva tools, which have rigged thousands of models for game creators.

bash

curl -X POST https://api.cinevva.com/v1/rigs \
  -H "Authorization: Bearer $CINEVVA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model_url": "https://example.com/goblin.glb", "rig_type": "biped" }'

json

{
  "id": "8f14e45f-ceea-467a-9c1a-1f0d0e6b7a21",
  "object": "rig",
  "status": "queued",
  "poll_url": "/v1/rigs/8f14e45f-ceea-467a-9c1a-1f0d0e6b7a21",
  "estimated_seconds": 30
}

Poll poll_url until status is succeeded, then download model_url. The quickstart walks the whole loop in about five minutes.

What you can call today

Capability	How you use it	Billed
Rigging	`POST /v1/rigs`	Per rig
Model generation	`POST /v1/models`	Per model
Retargeting	`animation_url` on a rig, or free in your engine	Per clip, server-side only
Animation catalog	`GET /v1/animations`	Free
Asset library	`GET /v1/assets`	Free
MCP server	`https://api.cinevva.com/create/mcp`	Per underlying operation
Optimize, convert, inspect	Browser tools	Free

Optimization and format conversion run entirely in your browser today, so they are tools rather than endpoints. They are listed on the tools page with what each one does, and flagged honestly where an API does not exist yet.

What it costs

API access requires an active Standard or Pro subscription, and work is billed per operation against your credit balance (100 credits = $1.00).

Operation	Credits	Approx. USD
Rig, Fast engine	2	$0.02
Rig, Pro engine	31	$0.31
Retarget, per clip	13	$0.13

Browsing the catalogs and polling a job cost nothing. Full detail on the pricing page.

Where to go next

Start with the quickstart if you want working code, or how rigging works if you would rather understand the pipeline first. Read model requirements before you send production assets: most failed rigs are input problems, and they are predictable.