PRACTICAL GUIDE · PROMPTTOVISUAL
Grok Imagine Video 1.5 Lite: Pricing, 1080p Upscaling and API Choices
Grok Imagine Video 1.5 Lite makes short videos with generated audio from text or one starting image. Output starts at $0.02 per second at 480p; 1080p is upscaled from 720p. This external model guide compares xAI and Runware. Lite is not integrated into PromptToVisual.
EXTERNAL MODEL GUIDE · CHECKED OCTOBER 9, 2026
Read xAI’s Lite model card → · Read the Runware contract →
No Lite generation is available in PromptToVisual. Published prices below include video output only; a starting image adds $0.01.
| Resolution | Per second | 10 seconds |
|---|---|---|
| 480p | $0.02 | $0.20 |
| 720p | $0.03 | $0.30 |
| 1080p (upscaled) | $0.14 | $1.40 |
Published estimates, not tested bills. See each channel’s pricing and input rules below.
What Lite supports, and when full 1.5 matters
xAI documents 1–15-second text-to-video and image-to-video output, with generated audio and 480p, 720p or 1080p delivery. Lite does not inherit full 1.5’s multiple image references, voice references, first-and-last-frame control or intermediate keyframes. Video editing and extension belong to the older grok-imagine-video model, not Lite. Choose the exact mode before comparing prices.
1080p delivery is not native 1080p generation
xAI explicitly describes Lite’s 1080p output as upscaled from 720p in both text-to-video and image-to-video modes. Full grok-imagine-video-1.5 supports native 1080p for those modes. A larger output frame alone does not establish more original scene detail. If fine text, faces or product edges matter, compare the actual exported result against your delivery requirements; this guide contains no independent output test.
xAI pricing: calculate output plus the starting image
Checked October 9, 2026: xAI lists USD $0.02/$0.03/$0.14 per output second at 480p/720p/1080p. Ten seconds therefore costs $0.20/$0.30/$1.40 for video output. One input image adds $0.01, making those image-to-video examples $0.21/$0.31/$1.41. These are calculations from published rates, not tested invoices or PromptToVisual credits. No promotional expiry is specified on the checked card. Verify current account and regional terms before paying.
Runware pricing: the same displayed units, a separate channel
Runware lists the same three per-second rates and a separate $0.01 input-image charge. Its six-second 720p first-frame example totals $0.19: $0.18 video plus $0.01 image. Every clip includes generated audio; the checked Lite schema has no audio-off control. A public catalog entry does not prove that a particular account can run the model. PromptToVisual has not completed Lite qualification or enabled it.
xAI API: model name and an asynchronous result
The direct model name is grok-imagine-video-1.5-lite. xAI’s documented REST flow submits to POST /v1/videos/generations and then polls GET /v1/videos/{request_id}. Use prompt, duration, resolution and aspect_ratio for text-to-video. Image-to-video follows the input image’s shape, so aspect_ratio is ignored. The generic guide defaults to eight seconds and 480p; specify both to avoid comparing different jobs. A returned request ID is not a completed video, and output URLs are temporary. No request was submitted for this guide.
Runware API: use its AIR identifier and its own fields
The AIR identifier is xai:grok-imagine@video-1.5-lite. Its schema uses taskType videoInference, a taskUUID and positivePrompt (up to 4,000 characters). For text-to-video, send a permitted width/height pair. For image-to-video, use exactly one inputs.frameImages item as the first frame and select resolution; do not manually send dimensions. Duration accepts integers 1–15, default six; resolution defaults to 480p. The response is asynchronous. These names and defaults are not interchangeable with xAI’s REST payload.
Sizing and directing a shot
Runware’s guide distinguishes text-only dimensions from first-frame resolution selection. For a text-only 720p landscape request, its example uses 1280 × 720. With a first frame, let the channel derive the shape from that image. Describe the subject, one clear motion, camera behavior, timing and sound in the prompt. A requested title or spoken line is an instruction, not a guarantee of exact text or lip synchronization. Use only images and audio-related material you have permission to provide.
Which route should you choose?
Our decision rule: consider Lite when a short text-led or single-image clip fits and per-second cost matters. Begin comparisons at the same duration and resolution; 1080p costs more even though Lite upscales it. If native 1080p, multiple references or pinned frames are essential, inspect full 1.5’s separate contract. Pick xAI for its direct interface, or Runware if your existing workflow already uses its task format and your account has access. Neither route was tested here. Vendor speed claims and sample timings are not an end-to-end service guarantee.
What can you use in PromptToVisual today?
Grok Lite video is not integrated here. The Grok image model on this site is a different service. PromptToVisual’s existing Seedance 2.5 option uses one permitted first frame and a prompt for a four-second video with audio, square or landscape, at 40 paid credits. It requires verified sign-in, eligible subscription or add-on credits, and runtime and safety eligibility. Trial credits do not qualify. That existing contract is unchanged; it does not offer Lite’s external pricing or longer clips. For a still-image text edit, use our current AI Image Editor.
Sources and next steps
Sources checked October 9, 2026. External model guide only. No integration, account qualification, generation test or measured benchmark. Provider rates are separate from PromptToVisual credits.
xAI capability comparison and 1080p distinction →
Runware exact Lite schema and rates →
Runware sizing and prompting guide →
Edit a still image with current PTV models · PTV Seedance alternative: four seconds, 40 paid credits · Compare Vidu channel contracts · All guides