Tries:
0
/
800
0%
MiniMax H3-Max Reference-to-Video
MiniMax H3-Max Reference-to-Video
Combine reference images, videos, and audio into one coherent video
MiniMax H3-Max Reference-to-Video
Multimodal references: up to 9 images (subject/style), 3 videos (motion, 2–15s each, ≤15s combined), and 3 audio clips (2–15s each, ≤15s combined) — 12 files max total.
Audio cannot be the only reference — always add at least one image or video with it.
Follows character identity, visual style, movement, and sound cues — reference assets positionally: Image 1, Image 2, Video 1, Audio 1, and so on.
Resolution: 480P / 768P. Prompt expansion: balanced (~1s) or quality (richer prompt, ~30s extra).
References & Prompt
Prompt
— Select a prompt —
Reference Images (up to 9)
Add Reference Images
Reference Videos (up to 3, 2–15s each, ≤15s combined)
Add Reference Videos
Cloned Voices (adds as reference Audio 1, Audio 2, …)
All Groups
— Select a voice to add —
Reference Audio (up to 3, 2–15s each, ≤15s combined)
Add Reference Audio
Settings
Resolution
768P (2K)
480P
Aspect Ratio
9:16
Adaptive
16:9
4:3
1:1
3:4
Duration (seconds)
Seed (optional)
Prompt Expansion Mode
Balanced (returns in ~1s)
Quality (richer prompt, ~30s extra)
Enable safety checker
Generate Video
Download
New Generation
Sessions
Session name
Save Session
New Session (Clear All)
Download / Export as File
Load / Import from File
Home
Info Page
Brainstorm & Build Your Prompt
Wipe conversation
×
Final Prompt