SpaceXAI: Grok Imagine Video 1.5

SpaceXAI: Grok Imagine Video 1.5 is a video generation model from xAI. It belongs to the video category and is identified on api14 as x-ai/grok-imagine-video-1.5. The provider release date is 2026-07-20. If you are building a pipeline that needs to turn written ideas or still images into short video clips, this is the type of model to evaluate.

On the input side, the model accepts text and images. Text prompts describe the scene, and an optional starting image can guide it, with first_frame supported as the frame image role. The output is video. It can render at 480p, 720p, or 1080p, in aspect ratios of 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, and 2:3. Clip durations run from 1 to 15 seconds in whole second steps, so you choose the length rather than receiving a fixed one. No context length is published for this model.

Those capabilities point to a few practical uses. A text to video workflow fits when you need a quick clip from a prompt alone, while supplying a first frame is useful when you already have a key visual and want motion that starts from it. The range of aspect ratios covers landscape, vertical, and square placements, which helps when the same clip must fit different screens. Short durations of a few seconds suit looping backgrounds, teaser shots, and social formats, while the longer end up to 15 seconds gives room for a fuller scene. Resolution choice lets you trade detail against speed depending on where the clip will be shown.

You call the model through api14's OpenAI-compatible API using the model id x-ai/grok-imagine-video-1.5. The supported request parameters are logprobs, max_tokens, response_format, seed, temperature, top_logprobs, and top_p. Because the interface follows the OpenAI format, you can point an existing OpenAI client at api14 and change the model id rather than rewriting your integration. The seed parameter is worth using when you want to reproduce or vary a generation across runs.

Note a few constraints before committing. The maximum output is listed as 0, so check the live model page for current output behavior rather than assuming a token budget. Audio generation is not specified, so do not plan on synchronized sound from this model. There is also no published context length, which matters if your prompts are long or reference large image payloads. If your task is text or image understanding rather than video creation, a model in those categories will fit better.