FRAMELOOP MODELS
Every Frontier AI Video Model, in One Studio
Browse the AI video generation models supported in Frameloop — see capabilities, watch sample videos, and start creating in a single canvas.
Alibaba
Happy Horse 1.0
The #1-ranked AI video model
1080p text-to-video and image-to-video with synchronized native audio, multilingual lip sync, and coherent multi-shot storytelling. Topped the Artificial Analysis Video Arena in April 2026.
- Joint video + audio in one pass
- 7-language lip sync
- Native multi-shot consistency
- 1080p, 3–15s clips
ByteDance
Seedance 2.0
Cinematic motion with native audio
Unified multimodal audio-video joint generation with industry-leading motion stability, physical realism, and dual-channel sound for cinematic output.
- Multimodal references (text, image, video, audio)
- Dual-channel stereo audio
- Director-level camera control
- Strong motion physics
More models coming soon. Want to see a specific model in Frameloop? Let us know.