Grok Imagine 2.0AI image generator
xAI's second Grok Imagine, with a 2K option and a cheaper draft quality. On Leap, it sits beside 174 other models in one studio.
- Text to image
- 7 aspect ratios
- No subscription
- Every top model, one account
- Pay per run, priced before it starts

AI image generator
Grok Imagine 2.0 at a glance
Grok Imagine 2.0 is an image model from xAI. It makes images from words, at up to 2K.
The version before it: Grok Imagine.
- Made by
- xAI
- Makes
- Images
- Starts from
- Words
- Resolution
- 1K or 2K
- Shapes
- 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3
- Per run
- Up to 4 images
- Prompt
- Up to 8,000 characters
- Price per run
- $0.054 to $0.11
- Updated
- October 7, 2026
How it's made
From a still to a film, in one place.
"Four on a scooter", replayed step by step: Grok Imagine 2.0 makes the still, Seedance 1.5 Pro brings it to life. Two models, one studio.


- 01Grok Imagine 2.0Makes the stillDone
Vertical night street photograph from a moving scooter alongside, 50mm lens, 1/30 s, rain streaks and light trails. A Hanoi street in a monsoon downpo...
- 02Seedance 1.5 ProBrings it to lifeDone
The family rides through the monsoon downpour alongside the camera; the father keeps the tower of red plastic stools perfectly balanced on his shoulde...
"Four on a scooter", made on Leap: two models, one studio, one library.
Made on Leap
More from Grok Imagine 2.0

Gol
Candid photograph, 35mm lens from beside the television, night. A packed corner bar in La Boca, Buenos Aires: brass beer taps, old framed football photos on the walls blurred, a crowd lit by the cold glow of the TV. A 45-year-old Argentine woman, curvy, dyed burgundy hair in a claw clip, gold hoop earrings, a blue-and-gold striped scarf with no crest, grips a glass of beer and leans toward the screen with clenched teeth and wide eyes in the moment before a goal; around her strangers of all ages hold their heads and an old man in a cap clasps his hands in prayer. Palette: blue and gold, TV glow, brass, beer amber. Grain, sweat, real faces. No readable text, no logos.

Northbound, shot 1
Extreme wide shot, 40mm anamorphic, blue hour. A single white Mongolian ger alone on a vast frozen steppe in Khentii province, smoke rising from its stovepipe, a few horses with frost on their manes standing nearby, low snowy hills. High in the darkening sky a thin streak of electric violet light crosses silently, trailing sparks. No readable text, no logos. Shot on 35mm anamorphic, ice fog, fine grain. Palette: ice white, felt brown, deel red and one electric violet. Cold overcast or blue-hour light.

Northbound, shot 6
Wide night shot, 24mm anamorphic: a white ger on the frozen steppe under stars, a herd of horses galloping through the snow in the foreground, and behind the ger a column of electric violet light rising silently into the sky. No readable text, no logos. Shot on 35mm anamorphic, ice fog, fine grain. Palette: ice white, felt brown, deel red and one electric violet. Cold overcast or blue-hour light.

Auntie's pepper sauce
Vertical selfie video frame from a front phone camera, slight wide-angle distortion, real phone-camera look. A Nigerian-British woman of 34, plus-size and round-faced, dark skin with a few dark spots on her cheeks, short natural hair in a tapered cut, gold hoop earrings, a faded oversized grey sweatshirt, stands in a small, slightly messy kitchen in a South London flat in the morning. She holds up to the camera a small glass jar of homemade red pepper sauce with a plain white handwritten label that says OJO, eyebrows raised, mouth open mid-sentence, playful and sceptical. Behind her: a teal kettle, a dish rack full of mugs, tea towels drying on a radiator, a window with flat grey London light and condensation. Natural side light from the window, slightly underexposed, no makeup look. Palette: sweatshirt grey, pepper red, kettle teal. Candid, unpolished creator content.

Rooftop legwork
Vertical documentary photograph, 28mm lens, dusk. A rooftop of rusted corrugated iron with a concrete water-tank stand on Lagos Island, Nigeria, overlooking dense rooftops, satellite dishes and the lagoon in violet haze; a yellow minibus in the street far below. A Nigerian man in his mid-forties, wiry and bald, a gap-toothed grin, sweat on his scalp, wearing a loose matching Ankara print shirt and trousers in cobalt blue and saffron yellow, white socks and black slides, mid-move in an Afrobeats legwork dance, one knee up, arms loose, shoulders bouncing. Two schoolgirls in green-and-white check uniforms sit on the water tank behind him, laughing and filming him on a phone. Light: last warm sun from the left, violet sky. Palette: cobalt, saffron, rust, dusk violet. Real sweat, dust and texture. No readable text.

Second line
Vertical parade photograph, 28mm lens from low in front of a marching drumline, afternoon sun. A street in Tremé, New Orleans, during a Sunday second line: shotgun houses, dancers with feathered parasols, a crowd. A 15-year-old Black snare drummer, slim, braces on his teeth, sunglasses pushed up on his head, a purple-and-gold high school band jacket unbuttoned over a white T-shirt, plays a rapid roll with his eyes closed and a huge grin, sticks blurred; behind him the drumline with bass drums and tenors recedes down the street. Palette: band purple and gold, brass shine, parasol pastels. Sweat, motion blur on the sticks, sun flare. No readable text, no logos.
Pick the right one
Grok Imagine 2.0 and its siblings
| Model | Best for | Resolution | Price per run |
|---|---|---|---|
| Grok Imagine 2.0 (this page) | xAI's second Grok Imagine, with a 2K option and a cheaper draft quality. | Up to 2K | $0.054 to $0.11 |
| Grok Imagine | Quick, vivid text to image at a low price. | – | $0.027 |
Side by side
More image models on Leap
No prompt needed
Or start from a recipe
Recipes write the prompt for you: add a photo, pick a look, and get a finished result.
Questions
Grok Imagine 2.0, answered
- What is Grok Imagine 2.0?
- Grok Imagine 2.0 is an image model from xAI. xAI's second Grok Imagine, with a 2K option and a cheaper draft quality. On Leap it runs in one studio beside every other top model, with one account.
- What resolution and sizes does it make?
- Up to 2K, in 1:1, 16:9, 9:16, 4:3, 3:4, 3:2 or 2:3.
- What's new in Grok Imagine 2.0?
- Grok Imagine 2.0 follows Grok Imagine. A default run costs $0.081, against $0.027 for Grok Imagine.
- How is it different from Grok Imagine?
- Grok Imagine 2.0: xAI's second Grok Imagine, with a 2K option and a cheaper draft quality. Grok Imagine: Quick, vivid text to image at a low price. They all run in the same studio, so you can try one prompt on each.
- How much does a run cost?
- A 1K image costs $0.081 at the default settings. By resolution and quality, a run costs $0.054 to $0.11. You see the exact price before you press go, and a run that fails costs nothing.
- Is Grok Imagine 2.0 free to try?
- No. There is no free credit: you add credit from $5, and your first top-up has no card fee. The cheapest run is $0.054 (a 1K image), so $5 covers 92 of those. You pay per run, in dollars, with no subscription.
- Can I use what I make commercially?
- Yes. You own what you make on Leap, so you can use it commercially, and we never train AI on it. The people and brands shown in a result can still have rights in it; the Terms say more.
- Is there an API?
- Yes. Leap's REST API runs it as xai/grok-imagine-image-2.0, at the same price per run as the studio, and agents can call it through Leap's MCP server. The request is at the bottom of this page.
- Do I need a xAI account?
- No. Leap sends your request to xAI's model and gives you what it makes. One Leap account runs Grok Imagine 2.0 and every other model, in the studio, through the API and from your agents.
Your idea, Grok Imagine 2.0.
Every run shows its price before it starts, and one that fails costs nothing.
Make it with Grok Imagine 2.0For developers
Run Grok Imagine 2.0 from code
- Model ID
- xai/grok-imagine-image-2.0
- Makes
- Image
- Pricing
- Per run, quoted first: POST /v1/quotes
Inputs
What the model takes. Leave out anything with a default.
The request
The same envelope for every model; only the model and its input change. With prefer: wait=60 the answer holds the result when the run finishes within a minute; a longer one is read again by its ID. Agents can call it through the MCP server at app.tryleap.ai/api/mcp.
curl https://api.tryleap.ai/v1/generations \ -H "x-api-key: $LEAP_API_KEY" \ -H "content-type: application/json" \ -H "prefer: wait=60" \ -d '{ "model": "xai/grok-imagine-image-2.0", "input": { "prompt": "Sports documentary photograph on a 200mm telephoto lens from low on the road, dawn. A red laterite dirt road between tea bushes and maize fields at 2,400 metres in Iten, Kenya, mist lying in the valley behind. A 16-year-old Kalenjin girl, very slim and long-legged, close-cropped hair, a faded maroon tracksuit top tied around her waist over a white vest, worn cheap trainers, runs straight at the lens leading a pack of six older male runners in mismatched training gear, her face set and calm, mouth slightly open, breath visible in the cold air. Low sun behind them backlights the red dust kicked up by their feet. Telephoto compression stacks the runners close together, shallow focus on her. Palette: red earth, mist blue, maroon. Real skin, sweat, dust on legs, slight motion blur on the feet. No race numbers. No readable text, no logos." } }' # Video often outlasts the wait and answers "queued" or "running".# Put its "id" in GENERATION_ID and read it again until it is done:# curl https://api.tryleap.ai/v1/generations/$GENERATION_ID -H "x-api-key: $LEAP_API_KEY"






