Grok Imagine 1.5: Features, Lite vs 1.5, and How to Use It

Grok Imagine 1.5 turns text, images or references into video with native audio. Compare 1.5 vs Lite, API prices, and how to use both on AI Studio.

A dark cinematic illustration showing a film frame being formed from glowing light particles with a sound waveform beneath it, in teal and warm orange
Table of contents

Grok Imagine 1.5 is a video generation model from SpaceXSI (formerly xAI) that creates clips up to 15 seconds long with synchronized native audio, dialogue, and sound effects. Available in two versions—full Grok Imagine 1.5 for native 1080p and multi-image references, and Grok Imagine 1.5 Lite for lower-cost drafts—it supports text-to-video, image-to-video, and character-consistent reference workflows.

You can run both Grok Imagine 1.5 and 1.5 Lite in your browser right here on AI Studio through the AI video generator. AI Studio is an independent website that runs several video and image models—including Seedance, Kling, Gemini Omni, Happy Horse, MiniMax H3, and Grok Imagine—on a credit system, and is not affiliated with SpaceXSI or xAI.

What is Grok Imagine 1.5 and when was it released?

Grok Imagine 1.5 is the flagship video and audio generation family from SpaceXSI, the company formerly known as xAI (which briefly became SpaceXAI in July 2026 before adopting SpaceXSI in October 2026, while keeping x.ai as its website).

SpaceXSI rolled out the 1.5 generation in two distinct phases during mid-2026:

  • June 16, 2026 (Initial Launch): SpaceXSI released grok-imagine-video-1.5 on the Imagine API alongside "Video 1.5 Fast" on grok.com/imagine and the iOS and Android apps. At launch, it worked strictly as an image-to-video model. The release focused on generating sound effects, ambience, and dialogue in the same pass as the video, delivering clearer lip-synced speech and stronger physical motion with fewer visual warps and more believable weight and momentum. In SpaceXSI's app, Video 1.5 Fast cut the generation time for a 6-second 720p clip down to about 25 seconds, compared to over 40 seconds on the previous version, and introduced workspace features like Projects, parallel agents, and library search.
  • July 31, 2026 (References & Text-to-Video Update): SpaceXSI expanded the model with the "Imagine Video 1.5 with References" update. This release added direct text-to-video generation without requiring a starting image, native 1080p resolution for both text-to-video and image-to-video, and reference capabilities. Users gained the ability to pass multiple reference images—each locking a specific subject like a face, a product, or a location—along with voice references to keep the same face and voice across separate scenes.

If you have seen mentions of a "2.0" model, note that no Grok Imagine Video 2.0 has been released or announced as of November 2026. "Grok Imagine Image 2.0" is a separate still-image model, leaving 1.5 and 1.5 Lite as the current video lineup. Furthermore, older API tasks like video editing and video extension remain on the legacy grok-imagine-video model rather than the 1.5 series.

What can Grok Imagine 1.5 do?

Grok Imagine 1.5 combines visual generation and audio synthesis into a single pass, supporting clips from 1 to 15 seconds in the API (and 4 to 15 seconds on AI Studio) at 480p, 720p, and 1080p resolutions. Because capabilities differ slightly between SpaceXSI's API and third-party browser interfaces, it helps to look at each feature individually.

Native audio and lip-sync

Both Grok Imagine 1.5 and 1.5 Lite generate synchronized audio automatically with every clip. Instead of bolting on background music after the video renders, the model produces ambient room tone, physical sound effects matching on-screen movement, and spoken dialogue with lip-sync in the same pass. When a character speaks in the prompt, the mouth movements and vocal delivery align with the action on screen.

Text-to-video

Added in the July 2026 update, text-to-video lets you generate clips from a written prompt alone. According to SpaceXSI's developer documentation, text-to-video on both 1.5 and 1.5 Lite works under the hood by first rendering a still frame from your text prompt and then animating that frame over your chosen duration. On the full 1.5 model, text-to-video renders up to native 1080p; on 1.5 Lite, 1080p output is upscaled from 720p.

Image-to-video

Image-to-video takes a single uploaded still image, sets it as the opening frame of the video, and animates the scene forward based on your text instructions. In both SpaceXSI's API and on AI Studio, image-to-video automatically preserves the aspect ratio of your uploaded picture.

References and voices

Reference generation allows you to supply visual or vocal anchors without forcing any single image to act as the literal first frame of the clip. Each reference image locks one element—such as a character's face, a specific product bottle, or a room layout—so you can place those elements into brand-new camera angles and actions.

  • Image references: Supported on full Grok Imagine 1.5 only (not on 1.5 Lite) at up to 720p for clips up to 15 seconds. While the July 2026 announcement introduced up to 7 references per generation, SpaceXSI's developer docs now support up to 14 reference images in the API. On AI Studio, reference-to-video supports 2 to 7 reference images at 480p or 720p.
  • Voice references (API and SpaceXSI app only): The API supports up to 3 voice references per clip to maintain the same voice across scenes, using preset voices or—for trusted partners on request—custom uploaded audio. Voice references are not available on AI Studio.

First/last frame and keyframes

In SpaceXSI's API and own tools, the full grok-imagine-video-1.5 model supports structural frame pinning that is not supported on 1.5 Lite or on AI Studio:

  • First and last frame: You can define both the starting image and the ending image of a clip. Passing the exact same image as both the first and last frame creates a loop.
  • Keyframes: You can pin up to 4 mid-video frames to direct where the motion must land at specific points in the timeline.

Grok Imagine 1.5 vs 1.5 Lite: which should you use?

Choosing between Grok Imagine 1.5 and Grok Imagine 1.5 Lite comes down to whether your shot requires multi-image character consistency and native 1080p sharpness, or if you just need a quick text-to-video or single-image animation at a fraction of the cost.

FeatureGrok Imagine 1.5 (grok-imagine-video-1.5)Grok Imagine 1.5 Lite (grok-imagine-video-1.5-lite)
1080p ResolutionNative 1080p (T2V and I2V)Upscaled from 720p
Available Resolutions480p, 720p, 1080p480p, 720p, 1080p
Duration1–15 s in API (4–15 s on AI Studio)1–15 s in API (4–15 s on AI Studio)
Native Audio & Lip-SyncIncluded on all clipsIncluded on all clips
Reference ImagesUp to 14 in API; 2–7 on AI Studio (capped at 720p)Not supported
Voice ReferencesUp to 3 voices (API / SpaceXSI app only)Not supported
First/Last Frame & KeyframesYes, plus up to 4 mid-video keyframes (API only)Not supported
API Price per Second$0.08 (480p), $0.14 (720p), $0.25 (1080p)$0.02 (480p), $0.03 (720p), $0.14 (1080p)
AI Studio Credits per Second6 (480p), 10 (720p), 18 (1080p)1.5 (480p), 2.5 (720p), 10 (1080p)
Best ForMulti-reference scenes, recurring characters, native 1080p outputLow-cost drafts, simple T2V/I2V clips, 480p/720p social videos

SpaceXSI's documentation summarizes the split clearly:

  • Pick Grok Imagine 1.5 Lite if you are making standalone text-to-video or single-image-to-video clips, testing prompt wording, or rendering at 480p or 720p where Lite costs about a quarter of the full model's credits on AI Studio.
  • Pick Grok Imagine 1.5 if you need to combine 2 or more reference images to keep a character, product, or setting consistent across shots, or if you want true native 1080p rendering instead of an upscale from 720p (or need pinned frames and voices via the API).

How much does Grok Imagine 1.5 cost?

Grok Imagine 1.5 pricing depends on whether you are billing per second as a developer through SpaceXSI's API or spending credits per clip inside AI Studio.

SpaceXSI API pricing

In SpaceXSI's developer documentation, video generation is billed per second of generated output according to resolution:

  • grok-imagine-video-1.5: $0.08 per second at 480p, $0.14 per second at 720p, and $0.25 per second at 1080p. Image inputs are also billed at about $0.01 per image.
  • grok-imagine-video-1.5-lite: $0.02 per second at 480p, $0.03 per second at 720p, and $0.14 per second at 1080p.

AI Studio credit rates

AI Studio does not charge in dollars per second; instead, every generation deducts credits based on the model, resolution, and duration you select. Total credits equal the per-second rate multiplied by the clip length in seconds, rounded up to the nearest whole number. Uploading first-frame images or reference images does not add extra credit charges.

Model on AI Studio480p (credits / sec)720p (credits / sec)1080p (credits / sec)
Grok Imagine 1.5 Lite1.52.510
Grok Imagine 1.561018

Here is what common clip lengths cost in practice on AI Studio:

  • 5-second clip at 720p (default settings): 13 credits on 1.5 Lite (12.5 rounded up) or 50 credits on full 1.5.
  • 10-second clip at 1080p: 100 credits on 1.5 Lite or 180 credits on full 1.5.
  • 15-second max-length clip at 1080p: 150 credits on 1.5 Lite or 270 credits on full 1.5.

You never have to calculate these totals by hand—AI Studio shows the exact credit cost before you click Generate Video.

How to use Grok Imagine 1.5 on AI Studio

You can generate videos with both models directly in your browser without managing API keys. Follow these steps:

  1. Open the generator: Go to the AI video generator on AI Studio. There is no separate standalone page for Grok; all models live inside the main video workspace.
  2. Select the model family: Open the model switcher and choose Grok Imagine Video.
  3. Pick your model version: Choose between Grok Imagine 1.5 Lite (listed first as the family default) and Grok Imagine 1.5.
  4. Choose a generation mode:
    • On 1.5 Lite, select Text to Video or Image to Video (upload 1 image to act as the opening frame).
    • On full 1.5, you can also select Reference to Video and upload between 2 and 7 reference images (such as a character portrait, an object, and a background).
  5. Write your prompt: Enter your scene description, camera movement, lighting, and any spoken dialogue or sound cues.
  6. Set duration, resolution, and aspect ratio:
    • Duration: Choose 4 to 15 seconds in 1-second steps (defaults to 5 seconds).
    • Resolution: Pick 480p, 720p, or 1080p (defaults to 720p). Note that if you are using Reference to Video on full 1.5, resolution is capped at 720p.
    • Aspect Ratio: In Text to Video, 1.5 Lite offers 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, and 2:3 (default 16:9), while full 1.5 offers 16:9, 9:16, 1:1, 3:2, and 2:3. In Image to Video, the output automatically matches your uploaded image's aspect ratio.
  7. Check the credit total and generate: Check the credit cost shown before you generate, then click Generate Video. Every finished video automatically includes native audio—there is no sound on/off toggle needed.

What AI Studio doesn't offer for Grok Imagine

To keep expectations clear, several developer-level or app-specific features exist only in SpaceXSI's own app or API and are not available on AI Studio:

  • Voice references and preset voice selection
  • First-and-last-frame pinning and mid-video keyframes
  • More than 7 reference images in a single prompt (the API allows up to 14)
  • Ultra-wide 21:9 and 5:2 aspect ratios
  • Clips shorter than 4 seconds (the API goes down to 1 second)
  • Video editing and video extension (which run on SpaceXSI's older grok-imagine-video API model)
  • SpaceXSI app workspace tools such as Projects and parallel agents

Prompt tips for Grok Imagine 1.5

SpaceXSI's developer documentation outlines a few core principles for getting reliable motion and clear audio out of the 1.5 architecture:

  1. Specify camera, lighting, and sound together in text-to-video: Because text-to-video renders an initial frame from your words and then animates it alongside the audio track, give the model concrete visual and sonic anchors. Name the shot framing, how the camera moves, the quality of light, and the exact ambient noises or dialogue you want to hear.
  2. In image-to-video, describe the motion—not the picture: When you upload a starting image, that picture already defines the opening frame, subject appearance, and aspect ratio. Do not waste your prompt re-describing what is already visible in the photo. Focus entirely on what changes after frame one: how the subject moves, how the camera tracks, and what sounds occur.
  3. Assign one clear job to each reference image: When using Reference to Video on full Grok Imagine 1.5, upload distinct images for each element you want locked—for example, Reference 1 for the character's face, Reference 2 for a coffee mug they are holding, and Reference 3 for the diner interior. Because references do not lock the first frame, use your text prompt to stage how those elements interact in the shot.
  4. Keep spoken lines paced to your clip duration: Native lip-sync works best when the spoken line naturally fits the seconds available. For a 5-second clip, write a single short sentence of dialogue rather than a long paragraph so the voice does not rush or get cut off at the end of the timeline.

Example text-to-video prompt (camera, lighting, and audio)

Medium tracking shot following a watchmaker in a leather apron walking past a wooden workbench at night. Warm tungsten desk lamps cast soft amber reflections across brass gears and glass domes. The camera slowly pushes in as he sits down, picks up a silver pocket watch, and looks up at the lens, saying calmly, "Every second leaves a mark." Crisp ticking of wall clocks, quiet footsteps on floorboards, and warm room ambience.

Example image-to-video prompt (motion and sound only)

The motorcyclist lowers her tinted visor with her right hand and twists the throttle. The headlight beam brightens against the wet asphalt as the bike launches forward out of the frame, kicking up a fine spray of rainwater. Camera pans right to follow the departure. Deep engine ignition roar, sharp visor click, and tires hissing on wet pavement.

Whether you want to draft quick 720p clips with Grok Imagine 1.5 Lite for a handful of credits or lock a recurring character across scenes using reference images in full Grok Imagine 1.5, having both models in one browser switcher makes it easy to match the tool to the shot. Head over to AI Studio's video generator, pick the Grok Imagine Video family, and try your first prompt with native audio.

Sources

Frequently asked questions

When was Grok Imagine 1.5 released?

SpaceXSI (formerly xAI) launched Grok Imagine Video 1.5 on June 16, 2026, initially as an image-to-video model on the Imagine API alongside Video 1.5 Fast on grok.com and mobile apps. A major update on July 31, 2026 added text-to-video generation, native 1080p resolution, image references, and voice references for keeping characters consistent across scenes.

What is the difference between Grok Imagine 1.5 and 1.5 Lite?

Grok Imagine 1.5 renders native 1080p video and supports reference images, voice references, first-and-last frames, and mid-video keyframes in the API. Grok Imagine 1.5 Lite is a lower-cost model for text-to-video and single-image animation; its 1080p output is upscaled from 720p, and it does not support reference images, voice references, or pinned keyframes.

How much does Grok Imagine 1.5 cost?

In SpaceXSI's API, grok-imagine-video-1.5 costs $0.08 per second at 480p, $0.14 at 720p, and $0.25 at 1080p, plus about $0.01 per input image. The Lite model costs $0.02 at 480p, $0.03 at 720p, and $0.14 at 1080p. On AI Studio, generation uses credits: 1.5 to 10 credits per second for Lite, and 6 to 18 for 1.5.

Can I use Grok Imagine 1.5 without a Grok subscription?

Yes. Developers can call the models directly through SpaceXSI's Imagine API and pay per second of output. If you want to generate clips in a browser without setting up API keys or a Grok account, you can use both Grok Imagine 1.5 and 1.5 Lite on independent platforms such as AI Studio using credits.

Is there a Grok Imagine Video 2.0 yet?

No. As of November 2026, SpaceXSI has not released or announced Grok Imagine Video 2.0. The current video models in the lineup are Grok Imagine 1.5 and Grok Imagine 1.5 Lite. While SpaceXSI does have a model named Grok Imagine Image 2.0, that release is strictly for generating still images, not video clips.

Create with AI Video Studio

Turn your idea into a video

Explore the video generator and start creating.

Generate a video

Written by

AI Studio Team