xAI Docs release notes

Video / Grok Imagine

Documentation now centers on grok-imagine-video-1.5 (examples, API samples, and model links updated from grok-imagine-video).

Documented modalities for grok-imagine-video-1.5:

• Text-to-video, image-to-video, and reference-to-video
• Native 1080p for text-to-video and image-to-video (reference-to-video capped at 720p)
• Generated videos include an audio track by default

Reference-to-video

• Guides generation with reference images, a preset voice, or both
• New Reference audio docs: up to 3 preset voices via reference_audios / voice_id (same roster as Text to Speech); tag in prompts as <AUDIO_0>…<AUDIO_2>
• Custom audio uploads are not supported
• Availability note: reference audio is currently United States only, for trusted partners

Related doc updates

• Imagine overview: “Image-to-Video” section retitled Video Generation; reference-to-video no longer described as unsupported on 1.5
• Image-to-video: optional prompt; 1080p called out; links to the 1.5 model page and Videos API
• Models index, cost-tracking, Files inputs/outputs, video extension, and REST videos reference examples point at grok-imagine-video-1.5

Release notes

• Added grok-imagine-video-1.5 modalities entry summarizing T2V / I2V / reference-to-video (including optional preset voices) and 1080p for T2V/I2V, with links to the video guides

Pages

• No pages added or removed