xAI Reference Mode supports up to 7 images, 3 preset voices, 1–15s duration, and 480p/720p resolution.
Drag images here or upload up to 7. Drag a thumbnail to reorder its <IMAGE_n> tag. Index 0 is the primary reference used for derived aspect ratio. Tag in prompt using <IMAGE_1>, <IMAGE_2>.
Select in order; the first image is primary.
Up to 7 images per batch.