The general-purpose multimodal route from MiniMax.
Text, image, video and audio context; T2V, I2V, reference generation and editing.
- Output768P or 2K · 4–15 seconds · native stereo audio
- See the route comparison →
Independent research directory · 04 Sep 2026
AI Live Stream maps MiniMax H3 Max routes, live-avatar platforms and public experiments into a compact, source-backed index for creators and developers.
Start with a job
Use the route that matches the question in front of you. The directory stays narrow so each entry can retain its source and operational limits.
Model reference
API IDs, provider differences, output limits and a dated source trail.
Curated index
Filter video APIs and real-time avatars without blending their workflows.
Practical tool
Estimate fal output cost from seconds, resolution and clip count.
Model desk
Text, image, video and audio context; T2V, I2V, reference generation and editing.
MiniMax documents text-to-video and image-to-video with first/last-frame controls.
Editorial boundary
We do not host generated videos or claim that a short-clip API is a continuous live-stream engine. Public project descriptions remain attributed to their authors.
Primary model source · Open reference · checked 04 Sep 2026