| Model | Price | Resolution | Duration | Native audio | Input |
|---|---|---|---|---|---|
| Grok Imagine Video 1.5 | 480p $0.098/s · 720p $0.166/s (+$0.015/img) | 480p / 720p | 1-15s | ✅ native | image-to-video (1 ref image) |
| Seedance 2.0 | 480p $0.092/s · 720p $0.197/s · 1080p $0.492/s | 480p / 720p / 1080p | 4-15s | ✅ optional | text / image / multimodal (≤9 img, ≤3 vid, ≤3 audio) |
| Seedance 2.0 Fast | 480p $0.071/s · 720p $0.153/s | 480p / 720p | 4-15s | ✅ optional | text / image / multimodal (faster & cheaper) |
| Kling Motion Control | 720p $0.06/s · 1080p $0.10/s (Kling 2.6) | 720p / 1080p | up to 30s | — | reference image + video (motion transfer) |
All four are callable through the apimodels.app unified video API: one API key, POST /api/v1/video/generations, 60-95% cheaper than official, reachable anywhere, $1 free on signup.
By per-second rate, Seedance 2.0 Fast is cheapest (480p $0.071/s). Kling Motion Control is $0.06/s at 720p but is motion-transfer (different use case). For general image-to-video, Seedance 2.0 Fast offers the best value.
Grok Imagine Video 1.5 topped the blind Image-to-Video Arena (720p), ahead of Seedance 2.0 — lifelike motion, strong prompt adherence, plus native synchronized audio (dialogue / SFX / ambience). Best for quality and finished-feel output.
Grok Imagine Video 1.5 generates synchronized audio alongside the video; Seedance 2.0 / Fast have optional generate_audio; Kling Motion Control has no audio.
Use Seedance 2.0: it accepts up to 9 reference images, 3 reference videos and 3 reference audios combined, plus 1080p output and up to 15 seconds.
Sign up at apimodels.app for one API key and call them all via POST /api/v1/video/generations with the model name (grok-imagine-video-1.5 / seedance-2.0 / seedance-2.0-fast / kling-motion-control) — 60-95% cheaper than official, directly reachable from anywhere.
