Gemini Omni Flash leads on human preference — 1370 vs 1325 Elo (#1 of 83 in the Image-to-Video Arena). The right pick still depends on style, speed and licensing — details below.
Gemini Omni Flash✓google/gemini-omni-flash/image-to-videovs
Grok Imagine Video 1.5 Reference to Videoxai/grok-imagine-video/v1.5/reference-to-video| Human-preference Elo | 1370 | 1325 |
| Arena rank | #1 of 83 | #6 of 83 |
| Category | image-to-video | image-to-video |
| License | commercial | commercial |
By human preference, Gemini Omni Flash currently rates higher (1370 vs 1325 Elo). Preference votes measure typical output quality; test both on your own prompts for style fit.
Both are image to video models available as hosted API endpoints — Gemini Omni Flash (google/gemini-omni-flash/image-to-video) and Grok Imagine Video 1.5 Reference to Video (xai/grok-imagine-video/v1.5/reference-to-video).
See the full best image-to-video models ranking.