The Content Array and Role System
- →Choose a generation mode (t2va, i2va, r2va) by constructing the right content array, not by hunting for a mode parameter
- →Label every media item with the correct role and predict what happens when a role is omitted
- →Explain why mixing frame roles with reference roles returns a 400, and recognize that error body on sight
If you go looking for a mode parameter in the H3 request schema, you will not find one. There is no "mode": "i2va", and — despite what most circulating guides say — no task_type field to set either. What you get instead is a single content array on POST /v2/video_generation, and the shape of that array is the entire mode-selection mechanism. Each element is distinguished by type (text / image_url / video_url / audio_url) and can be labeled with a role. One rule is absolute: every request must include one non-empty text item — the prompt is required — otherwise you get a parameter error before anything else is even considered.
The documentation defines the supported combinations, and they map exactly onto the three modes you met in Foundation:
Unlock “The Content Array and Role System”
The 4-lesson Foundation module is free and always will be — you have already read the part most guides get wrong. The paid lessons go deeper, with labs, knowledge checks, and every technical claim verified against live documentation.
Already bought it? Sign in with the email you used at checkout.