Which generation modes are available?
Image to Image, Text to Image, and Image to Video. Image to Video can generate voice audio first and then proceed to the video render from one job.
Short answers about the first MagicPixels release.
Image to Image, Text to Image, and Image to Video. Image to Video can generate voice audio first and then proceed to the video render from one job.
All prompt and voice text fields are limited to 500 characters.
New accounts include 1 free credit. Image jobs cost 1 credit. Generated audio costs 1 credit per 10 seconds. Talking-avatar video costs 22 credits per second.
Yes. Create an API key from a paid account, then use the REST API or MCP JSON-RPC endpoint.