Grok Imagine Image 2.0AI Image Generation and Editing by xAI
Grok Imagine Image 2.0 is an AI image generation and editing model from xAI. Generate from a prompt or edit up to three reference images with plain language, at 1K or 2K, in 13 aspect ratios, priced from $0.04 per image.
What makes Grok Imagine Image 2.0 different
Text that reads inside a generated image
Grok Imagine Image 2.0 renders text as part of a scene rather than pasting it on top of one, which is what makes it usable for anything with words on it.
- Holds a dozen separate blocks of text in one generation, at different sizes and on different surfaces.
- Type takes on the material it sits on: ink settling into leather grain, print curving with the form, letters breaking over a crease.
- Spell the wording you want in the prompt, one line per line of the lockup, and the hierarchy comes back as written.
- Built for product prints, packaging, garment graphics, signage, and campaign art direction.
Edit images with plain language, not masks
The xai/grok-imagine-image/v2.0/edit endpoint takes your images plus a written instruction and returns the change with the rest of the frame preserved, with no masks, no layers and no selection tools.
- Up to three input images per request.
- Recolor a product, rewrite the type printed on it, change a location, or combine subjects from separate images.
- Feed the result straight into Grok Imagine Video 1.5 as a first frame to put the same scene in motion.
- The default
autoaspect ratio keeps the framing of your first input image. - Input images add $0.01 each on top of the output price.
Two image quality modes at 1K or 2K
A single quality parameter lets you decide per request how much fidelity an image is worth, which is new in the 2.0 generation of Grok Imagine.
- Low or medium quality, at 1K or 2K resolution, for four price points between $0.04 and $0.08 per image.
- Up to four images per request from a single prompt.
- 13 aspect ratios, from 2:1 ultrawide through square to 9:20 vertical.
- JPEG, PNG, or WebP output.
How much Grok Imagine Image 2.0 costs
Both endpoints bill per output image, and the rate follows the quality mode and the resolution you ask for. For the rest of the family, see Grok Imagine for every endpoint and rate, Grok Imagine Video 1.5 for video, and Grok Imagine Free for the free daily tier. Pricing information updated as of August 13, 2026.
Grok Imagine Image 2.0 API endpoints
Generate from a prompt or edit existing images, both running on the same 2.0 model.
What are examples of Grok Imagine Image 2.0 output?
One campaign, one bag, four setups: open field, service corridor, stairwell, and a frame with nothing in it but sky. Each is a single Grok Imagine Image 2.0 pass at 2K, no photographer, no studio, no retouching. Copy any prompt and run it yourself.
Dead field, 28mm from the ground
"Fashion campaign photograph shot on 35mm film with a 28mm lens from a very low camera position looking slightly up, hard single-source on-camera flash, cold and frontal. A lean young East Asian man with a bleached blonde crop crouches low in dead straw grass, forearms on his knees, gaze cast down and away from the camera, a black leather duffel bag set on the ground beside him with GROK IMAGINE and IMAGE 2.0 printed large and flat across it in baby pink capitals. Wide-angle proximity distorts the foreground grass. Flat white-grey overcast sky filling the top of the frame with no cloud detail, no horizon interest, no atmospheric depth. Crushed desaturated palette, near-black shadows, anti-glamour posture, visible 35mm grain. No lens flare, no haze, no rim light."
Service corridor, 135mm compression
"Fashion campaign photograph shot on 35mm film with a 135mm lens, compressed and flat, hard single-source on-camera flash. A lean young East Asian man with a bleached blonde crop walks away down a bare institutional corridor, small in the frame at the far end, a black leather duffel bag hanging from one hand, GROK IMAGINE and IMAGE 2.0 printed large and flat across it in baby pink capitals and catching the flash. Painted breeze-block walls, a scuffed vinyl floor, one closed door, dead overhead fluorescent light with no warmth. Long lens compression flattens the corridor into stacked bands, crushed desaturated palette of grey-green and gunmetal, near-black shadows, no atmospheric depth. Visible 35mm grain, late-1990s magazine editorial. No lens flare, no haze, no rim light."
Stairwell, straight down
"Fashion campaign photograph shot on 35mm film with a 35mm lens from directly overhead, looking straight down a concrete stairwell, hard single-source flash. A lean young East Asian man with a bleached blonde crop sits slumped on a bare concrete step far below, a black leather duffel bag across his knees, the top of his head and his shoulders foreshortened by the top-down angle, GROK IMAGINE and IMAGE 2.0 printed large and flat across the bag in baby pink capitals. Painted steel handrails and stair nosings make hard graphic diagonals across the frame. Crushed desaturated palette of concrete grey and gunmetal, near-black shadow in the well, no atmospheric depth, anti-glamour posture, visible 35mm grain. No lens flare, no haze, no rim light."
Mid-air, 24mm, flat sky
"Fashion campaign photograph shot on 35mm film with a 24mm lens from a low angle, hard single-source flash freezing the motion. A black leather duffel bag is caught mid-air, tumbling above a dead open field, handles flung out, GROK IMAGINE and IMAGE 2.0 printed large and flat across it in baby pink capitals, tack sharp against a flat white-grey overcast sky with no cloud detail. Far below at the bottom edge of the frame, a blurred figure in indigo denim, out of focus and cropped. The bag sits high and centred with enormous empty sky around it, crushed desaturated palette, near-black shadow inside the bag mouth, visible 35mm grain, strange and deadpan. No lens flare, no haze, no rim light."
How to use the Grok Imagine Image 2.0 API
The client API handles the request submit protocol. It will handle the request status updates and return the result when the request is completed.
import { fal } from "@fal-ai/client";
const result = await fal.subscribe("xai/grok-imagine-image/v2.0/text-to-image", {
input: {
prompt: "Fashion campaign photograph on 35mm film with a hard single-source frontal flash, a black leather duffel bag square to camera against a flat grey wall, heavy baby pink capitals across the leather reading 'GROK IMAGINE' then 'IMAGE 2.0' above a thin rule and three smaller lines, crushed desaturated palette, visible film grain",
aspect_ratio: "16:9",
resolution: "2k",
quality: "medium",
num_images: 1,
},
logs: true,
onQueueUpdate: (update) => {
if (update.status === "IN_PROGRESS") {
update.logs.map((log) => log.message).forEach(console.log);
}
},
});
console.log(result.data);
console.log(result.requestId);
Common questions about Grok Imagine Image 2.0
What is Grok Imagine Image 2.0?
Grok Imagine Image 2.0 is the newest generation of xAI's Grok Imagine image model, published on fal on August 11, 2026. It covers two jobs from the same model: text to image generation and instruction based image editing. Output runs at 1K or 2K resolution across 13 aspect ratios, with a quality parameter that sets the fidelity and the price of each request.
What is new in Grok Imagine Image 2.0 compared with earlier Grok Imagine image models?
Grok Imagine Image 2.0 is a new model generation, and it is the first Grok Imagine image model with a quality parameter, so a single endpoint covers cheap drafts and finished renders. The earlier tiers are still available on fal: the original Grok Imagine Image at $0.02 per image, and Grok Imagine Pro at $0.05 (1K) or $0.07 (2K) per image. All of them sit under the Grok Imagine family.
How much does Grok Imagine Image 2.0 cost on fal?
Grok Imagine Image 2.0 is billed per output image, with the rate set by quality and resolution: $0.04 per image at low quality 1K, $0.06 at medium quality 1K, $0.06 at low quality 2K, and $0.08 at medium quality 2K. Image editing charges the same output rate plus $0.01 for each input image you pass. There is no subscription and no minimum spend, so a first test costs a few cents.
Is there a free way to try Grok Imagine Image 2.0?
Grok Imagine Image 2.0 is billed per image with no free tier of its own, and low quality at 1K keeps a first test to $0.04 an image. If you want to try the family without paying per generation, Grok Imagine Free covers daily video generations in the fal sandbox, and Grok Imagine lists every endpoint in the family with its rate.
What is the difference between low and medium quality?
The quality parameter on Grok Imagine Image 2.0 accepts low or medium, and it changes both the fidelity of the render and the price of the request. Medium is the default and is the setting to use for finished work. Low costs $0.02 less per image at either resolution, which makes it the cheaper way to explore a prompt or generate a batch of four options before committing to a final render.
What resolutions and aspect ratios does Grok Imagine Image 2.0 support?
Grok Imagine Image 2.0 generates at 1K or 2K resolution and supports 13 aspect ratios: 2:1, 20:9, 19.5:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:19.5, 9:20, and 1:2. Text to image defaults to 1:1, while the edit endpoint adds an auto option that preserves the aspect ratio of the first input image. Each request returns between one and four images as JPEG, PNG, or WebP.
How does image editing work in Grok Imagine Image 2.0?
The xai/grok-imagine-image/v2.0/edit endpoint takes a written instruction plus up to three input images and returns the edited result. There are no masks, layers, or selection tools: you describe the change, and the model applies it while leaving the rest of the frame alone. Passing more than one image lets you combine subjects, for example placing a product from one photo into the scene of another. The response also includes the enhanced prompt the model actually used, in the revised_prompt field.
Can Grok Imagine Image 2.0 render readable text in images?
Yes. Grok Imagine Image 2.0 renders printed and screened type as legible text, and it holds a full typographic lockup at several sizes in a single generation. Every image on this page was generated with the model on fal, including a five line print across a leather bag and a macro of that print at 2K, close enough to see the ink sitting down in the grain of the hide. Write the exact wording you want inside the prompt, keep each line of the lockup on its own line, and the model will follow the hierarchy you describe.
Does Grok Imagine Image 2.0 generate video?
No. Grok Imagine Image 2.0 is an image model. Video generation in the Grok Imagine family runs on a separate model, Grok Imagine Video 1.5, which covers text to video, image to video, and reference to video with native audio. An image generated here can be passed straight into the image to video endpoint as a first frame.
How do I get started with the API?
Install the fal SDK (Python or JavaScript), grab an API key from your dashboard, and call xai/grok-imagine-image/v2.0/text-to-image with a prompt. The API is serverless, so there are no GPUs to manage and no infrastructure to set up. Check the API documentation for every available parameter.
Can I use Grok Imagine Image 2.0 for commercial projects?
Yes. Images generated through the fal API can be used in commercial projects. Check fal's terms of service for full details on usage rights and licensing.
Get in touch about Grok Imagine Image 2.0
Want to learn more about integrating Grok Imagine Image 2.0 into your workflow? Leave your details and our team will reach out.