MiniMaxAI/MiniMax-H3
Image-Text-to-Video • 33B • Updated • 4.04M • • 4.37k
MiniMax H3
-
MiniMax H3
🎬274Video generation with a synchronized soundtrack
Note Text-to-video, Image-to-video
MiniMax-H3 Reference
🎭85Unquantized MiniMax-H3 from image, audio, video refs
Note Reference to video (image-reference, audio-reference - including lipsync, video reference)
-
Qwen3-VL conditioner
🧠3Qwen3-VL 33B prompt conditioning as a service