Models › H3 Max
MiniMax H3 Max, image-to-video
minimax/h3-max/image-to-video
1 framenative audio5–15 s768p native
price$0.08/s
wait p50 / max7s / 13s
checks all green5 of 9
lines heard100%
refused0
spent here$5.20
Post-trained by fal for prompt adherence. One image in, no multi-reference. Continues the frame rather than re-staging it, which is why it holds identity better than models with more reference slots.
Everything it rendered
9 inputs, 11 takes. Click any one for the full request and its measurements.
Repeatability
The same input sent again, more than once. The number is the spread of motion across replicates over its mean — lower is steadier.
| Input | Replicates | Motion CV | Detail |
|---|---|---|---|
| Stylised animation | 3 | 0.164 | 3 replicates · motion 0.64 / 0.89 / 0.76 |
Identity across calls
| Character | Mean similarity | Appearances | Detail |
|---|---|---|---|
| The baker | 0.967 | 2 | 2 calls compared to their own centroid · 10/10 sampled frames passed the fidelity gate |
| serial-flyer | — | 1 | only one input produced a usable appearance, so there is nothing to compare across calls |
| The paper fox | 0.955 | 2 | 2 calls compared to their own centroid · 10/10 sampled frames passed the fidelity gate |