Start at 250 credits
The default is a 5-second 768p text-to-video draft. It is the lowest-cost MiniMax H3 run in AuraTuner and needs no upload.
Start with a 5-second, 768p text-to-video draft for 250 credits—no upload required. Use MiniMax H3 from a prompt, a first frame with an optional end frame, or a reference video when the shot needs tighter direction.
MiniMax H3 Studio
Start with text, a first frame, or source video. MiniMax H3 is already selected.
Use the smallest control that solves the shot. Start from text when the idea is open; add a frame or source clip only when it gives the model a real constraint to follow.
The default is a 5-second 768p text-to-video draft. It is the lowest-cost MiniMax H3 run in AuraTuner and needs no upload.
Add a first frame to lock the opening look; add an optional end frame when a reveal, transition, or final composition matters.
Use video-to-video when motion or timing from a source clip matters. It is more expensive, so reserve it for a constraint a prompt cannot express.
Use 768p to select direction. Move a chosen shot to 2K only when detail or crop room matters.
The input should already contain the part of the shot you cannot afford to lose: the idea, the opening look, or the motion.
Use a short shot brief when the scene, framing, and action are still open. Pick the delivery ratio before writing the composition.
Use an approved visual as the first frame when product shape, character, art direction, or opening composition must stay close.
Use a source clip when its movement or timing is the constraint. It costs more per second than text or image input.
Use an approved image as the first frame to lock the opening look. Add an end frame only when the final composition matters: a product reveal, transition, or deliberate closing shot.
One important constraint
An end frame is optional, but it needs a first frame. Use text-to-video when both compositions are still open; reserve two-frame direction for a shot that really needs both states controlled.
For the first 5-second draft, describe one subject action, one camera move, and one thing that must not change. The default Studio prompt follows this shape so you can edit it instead of starting from a blank field.
Name the product, person, or scene and its opening composition.
Ask for one action and one camera direction, not a montage.
State what must stay stable: shape, identity, framing, or text-free surfaces.
Yes. This page opens a 5-second 768p text-to-video starter that costs 250 credits and needs no upload. Switch to image-to-video or video-to-video when a frame or source movement must be preserved.
AuraTuner's available model on this page is MiniMax H3. Check the selected Studio model and supported controls above before starting a generation from a Hailuo 3.0 search.
MiniMax H3 is marketed with native-audio capability, but AuraTuner's current H3 Studio does not expose audio generation or audio-reference controls. The H3 runs on this page create silent video; add sound in a separate editing step.
AuraTuner supports text-to-video, image-to-video, and video-to-video for MiniMax H3. Choose the mode based on whether the prompt, still frame, or source motion is the constraint you need to preserve.
At 768p, text-to-video and image-to-video cost 50 credits per second, while video-to-video costs 100 credits per second. The 5-second starter on this page is 250 credits; 2K and video-to-video cost more.
Do not start with MiniMax H3 if you have not decided the source frame, motion, or output placement. First settle the still or run a smaller motion experiment, then use a narrow H3 test when the constraint is clear.