MiniMax Hailuo 2.3 variants & parameters
| Parameter | Hailuo 2.3 | Hailuo 2.3 Fast |
|---|---|---|
| Duration | 6 / 10s | 6 / 10s |
| Aspect ratios | — | — |
| Resolution | 768p | 768p |
| Native audio | ✓ | — |
| Image-to-video | ✓ | ✓ |
| Reference-to-video | — | — |
| Credits per second | 3 | 2 |
| 5-second clip cost | 15 cr/5s | 10 cr/5s |
What is MiniMax Hailuo 2.3?
MiniMax Hailuo 2.3 is an instruction-focused AI video model offered in Standard and Fast variants. The Standard model supports both text-to-video and image-to-video, while Hailuo 2.3 Fast is designed for lower-cost image animation when you already have a starting frame. Both variants use fixed 6- or 10-second clip lengths and support 768p output; 1080p is available for 6-second clips.
Hailuo's practical strength is direct camera language. It recognises 15 named moves in plain prompts, including zoom in and out, pan left and right, tilt up and down, dolly forward and back, push in, pull out, tracking, orbit, handheld, static, and crash zoom. This makes it useful when the movement of the shot matters as much as the subject.
The Standard model costs 18 credits for a 6-second 768p clip or 30 credits for 10 seconds. Hailuo 2.3 Fast costs 11 credits for 6 seconds or 18 credits for 10 seconds at 768p. The 1080p tier uses a 1.6 multiplier and is limited to 6 seconds. Exact pricing is shown in the generator before credits are spent.
Choose Standard when you are starting from text, need the full instruction-following workflow, or want to animate an image with the established Hailuo behavior. Choose Fast when you already have a source image and want more iterations for the same credit budget. For native 2K, flexible 5–15 second duration, first/last-frame interpolation, or image/video/audio references, use the separate MiniMax H3 model.
MiniMax Hailuo 2.3 vs MiniMax H3 vs Kling V3
| Capability | MiniMax Hailuo 2.3 | MiniMax H3 | Kling V3 |
|---|---|---|---|
| Native audio | |||
| Image-to-video | |||
| Reference-to-video | |||
| First/last frame |