Get started

Same account and commands from Node or Python. Agents: paste set up renda — https://renda.vochsel.com/llms.txt.

npx @rendahq/cli init            # trial or sign in, MCP + skills for your agent
npx @rendahq/cli render chair.blend --res 1920x1080

The Renda package

A small typed JSON file (*.renda.json) for what scene formats can't express: output, resolution, environment map, camera, animation, engine. Everything except scene.root is optional. JSON Schema: /api/v1/schema.json.

Examples

{
  "version": 1,
  "scene": {
    "root": "./model.glb"
  }
}

Uploading and rendering

# 1. register each file by sha256; PUT bytes if the server doesn't have them
curl -X POST $RENDA/api/v1/uploads -H "authorization: Bearer $KEY" \
  -d '{"filename":"robot.glb","size":48213,"sha256":"<hex>"}'
# → {"assetUri":"asset://<hex>","exists":false,"uploadUrl":"https://…"}
curl -X PUT --data-binary @robot.glb -H "content-type: application/octet-stream" "<uploadUrl>"

# 2. submit, using asset:// uris
curl -X POST $RENDA/api/v1/jobs -H "authorization: Bearer $KEY" \
  -d '{"package":{"scene":{"root":"asset://<hex>"},"output":{"format":"png"}}}'

# 3. poll until status is succeeded, then download outputs[].url
curl $RENDA/api/v1/jobs/<id> -H "authorization: Bearer $KEY"

The renda CLI and MCP server do the upload step for you. Full API: OpenAPI.

Fields