MiniMax Reference to Video + Audio (ref2va)

1-3 reference images, a tone/style reference audio, and a prompt in — a generated video clip with its own soundtrack out. Test it live with your own API key, or read how it works.

Also see:Text to Video (t2va)Image to Video (i2v)Lip Sync Video (ia2v)Music Video API

API key

Enter your API key to submit or inspect tasks.

Create a task

Fields map directly onto POST /aigc/task with task_type: "minimax_ref2va".

Model

Reference image 1Reference image 2 (optional)Reference image 3 (optional)

The first is required; the other two are optional extra references. Must be public http(s) URLs — this API does not accept base64.

Reference audio (tone / style only — not the final soundtrack)

This steers tone and style; it is not reused as the final audio and it is not lip sync.

PromptDescribe the scene and action; also describes the sound generated for the result.

Aspect ratioResolution

Duration (seconds)Integer, 3–15 seconds.

Estimated cost$0.0750

Client-side estimate using the current pricing formula — the actual charge may differ slightly with account discounts.

Or inspect an existing task

Request log

Every call this page makes to the API, in order, most recent first.

Nothing sent yet.