Limited Offer · 50% OFF50% OFFYearly plans · GPT Image 2.5 + Seedance 2.5
Image 2
Turn a detailed creative brief into a finished visual.
No reference image and no setup — the written brief is the whole input. Pick the ratio your work needs and the result comes back ready to use.
Nine original examples across editorial design, product concepts, fashion, campaigns, cinematic scenes, street photography, illustration, and surreal art. Select any image to reuse its prompt.
Move from a short idea to a carefully directed image by describing what matters and choosing the right canvas before generation.
Write the scene like a creative brief, from subject and materials to lighting, mood, and camera language.
Guide framing, visual hierarchy, placement, and any text that belongs inside the image.
Choose square, portrait, landscape, widescreen, or vertical output before you generate.
Move between editorial graphics, product photography, fashion, architecture, and surreal scenes without changing workflows.
One text box covers the visuals most teams need week after week.

Set the headline, the focal subject and the layout in one brief, then generate at 2:3 or 3:2 for print and event use.

Explore materials, finishes and lighting for a product idea long before anything is photographed.

Opening spreads, section dividers and article art that match the tone of a story instead of generic stock.

Square and vertical posts with a clear focal point and space left for captions in the app.

Mood frames, location studies and pitch visuals in 16:9 with the light and weather you describe.

Seasonal key visuals and pitch boards you can iterate on in minutes across several ratios.
Full-length briefs you can paste as they are, then swap the subject, materials, lighting or layout.

Vertical exhibition poster, 2:3. Deep charcoal ground with a single luminous amber glass ribbon curving from the lower left to the upper right, soft internal caustics where it folds. Museum-catalog grid: the headline "AFTERGLOW" set large in a modern serif across the top third, the subtitle "A STUDY IN LIGHT" in small letterspaced caps below it, a thin rule and a date line at the bottom margin. Generous negative space, gallery print finish.

Square studio packshot of an original translucent portable speaker on a dark stone pedestal. Smoked glass housing, brushed aluminium end caps, a warm light source glowing from inside the mesh. One large softbox from the upper left, a narrow strip highlight down the right edge, matte background falling to near black. Slight three-quarter top-down angle, razor-sharp product, subtle contact shadow, no branding.

Wide cinematic frame, 16:9. A small glass observatory standing alone on a rust-red desert plateau at blue hour, interior lights spilling warm gold across the sand. Distant mesas fade into haze along the horizon, the upper two thirds of the frame given to a deep twilight gradient with the first stars showing. Low camera position, anamorphic feel, fine film grain, no people and no vehicles.

Vertical editorial fashion photograph of a fictional adult model in an ivory architectural pleated outfit, standing inside a minimal curved concrete space. Diffused daylight from a hidden ceiling slot, soft falloff into shadow on the right, calm posture with relaxed hands. Medium-format look, shallow depth of field, natural skin texture, muted ivory and concrete palette, no logos and no readable text.

Square commercial still life of an unbranded frosted citrus drink can, condensation beading across the surface, suspended blood-orange slices and a burst of sparkling water frozen in mid-air around it. Vivid yellow acrylic planes stacked behind as a set, hard key light for crisp edges with a soft fill, saturated but clean color, high-speed capture feel, no brand marks.

Vertical surreal image, 9:16. A luminous koi swims inside a transparent glass cube floating a few metres above a moonlit mountain lake. The water in the cube glows faintly and casts rippling light onto the surface below, distant pine ridges silhouetted against a star-dense sky. Cool blue palette with a single warm accent from the fish, still air, dreamlike but photographic.
Three steps, no learning curve.
Describe the subject, environment, style, lighting, composition, and desired text.
Select the aspect ratio that matches where the image will be used.
Create the image, review the result, and improve the prompt when you want a different direction.
Answers to your questions about Grok Imagine Image 2.0.
Grok Imagine Image 2.0 is xAI's text-to-image model, available on Image 2. It turns a written creative direction — subject, setting, composition, lighting, style and any on-image text — into a finished image, without needing a reference photo.
You can generate in 1:1, 2:3, 3:2, 16:9 and 9:16, so the same prompt covers square social posts, portrait posters, landscape spreads, widescreen frames and vertical stories.
Not yet. The Image 2 launch of Grok Imagine Image 2.0 is text-to-image only, so uploads and image edits are disabled for this model. If you need reference-driven editing, use Nano Banana 2 or GPT Image 2 instead.
Yes. Put the exact wording in quotation marks in your prompt and say where it belongs — headline, subtitle, label — and keep the text short. Long paragraphs and dense small type are still the hardest case for any image model.
Each generation costs 3 Image 2 credits. You can claim daily check-in credits, buy a credit pack, or invite friends for more — see pricing for the current packs and plans.
This page generates still images from text. If you want motion, Grok Imagine Video 1.5 animates a single image into a short clip — a common workflow is to create the frame here first, then send it to the video model.
It depends on the Image 2 terms, the model's own terms, and your rights to the prompts, brands and references involved. It's worth reviewing these before you publish client or paid work.
Write the prompt like a brief rather than a keyword list: name the subject, the setting, the light, the materials, the camera or layout language, and the exact text you want. Then choose the aspect ratio before generating, and change one variable at a time when you iterate.
Your prompt and selected settings are processed to create the image, and successful results can appear in your Image 2 history for later access. Avoid including confidential or sensitive personal information in prompts. For details about data processing, storage and your choices, read our Privacy Policy.
Still have questions? Contact us at [email protected]
Turn your next creative brief into an image in the format you need.
Generated Images