Model Marketplace

Browse and access top AI models with unified OpenAI-compatible API format, ready to use.

Text to videoImage to videoReference to videoFirst/last frame

VIDEO

Open-weight omni-modal video model with joint audio-video output, supporting text-to-video, first/last-frame and omni-reference generation.

Mainstream Models

All Modalities

OpenAI Compatible

1 models

NEW

MiniMax-H3

Open-weight omni-modal video model with joint audio-video output, supporting text-to-video, first/last-frame and omni-reference generation.

VIDEO

Text to videoImage to videoReference to videoFirst/last frame

768P/s

¥0.5

Input image/img

¥0.2

Input video/s

¥0.5

VendorMiniMax