Grok Imagine Image Quality Text to Image API

xai/grok-imagine-image-quality/text-to-image

1K / 2K · 5 aspect ratios

Grok Imagine Image Quality Text to Image transforms text prompts into high-fidelity images up to 2K resolution, with advanced multilingual text rendering, photorealistic detail, and 5 aspect ratios. It adheres closely to complex optical and lighting prompts while preserving nuanced composition and natural texture throughout the scene.

Get API Key

Input

Prompt

973

Aspect Ratio

Resolution

OutputReady

Grok Imagine Image Quality generated result 1

Continue with

Grok Imagine Image Quality Edit Grok Imagine Image 2.0 Text to Image Grok Imagine Video 1.5

Examples

A photorealistic documentary photograph inside a basalt sea cave at low tide. One adult woman geologist with short dark hair, a weathered ochre field jacket and a small canvas backpack studies the columnar rock wall in three-quarter profile. Her right hand holds a small warm-white flashlight aimed at a nearby mineral seam; her left hand rests naturally on a closed field notebook at her waist. Cool blue daylight from the cave entrance on the left meets the restrained amber flashlight pool on the right. Believable wet basalt microtexture, small tidal reflections, natural skin pores and damp hair, fine sea spray in the distant opening. Medium-wide eye-level 35mm composition, person on the right third, coherent perspective and natural dynamic range, face and near rock in focus, distant ocean softer. Quiet field research, no theatrical posing, no additional people, no writing. No advertisement, brand, logo, fox, product display, promotional graphics, or watermark.

An intimate wildlife photograph of exactly one Japanese macaque resting beside the exposed roots of a cedar tree in a snowy mountain forest. The macaque is seated in a natural relaxed posture, three-quarter view, with both hands resting loosely in its lap and a small visible breath cloud in the cold air. Fine silver-brown fur with individual snow crystals, warm pink face, alert dark eyes, anatomically believable hands. Soft overcast morning light, cool white snow, warm natural fur, distant cedar trunks gently out of focus. Eye-level 85mm lens, shallow but credible depth of field, entire face sharply focused. Vertical composition with uncluttered negative space above, a candid moment without human objects or clothing. No text. No advertisement, brand, logo, fox, product display, promotional graphics, or watermark.

A cinematic yet realistic street-level architectural photograph of a small community library at night. Through a broad clear window, exactly two adult readers sit separately at a long wooden reading table beneath warm pendant lamps; shelves recede with consistent perspective. Outside is a quiet dry sidewalk and a deep blue evening sky, no rain. Beside the entrance, one simple illuminated rectangular wayfinding sign displays exactly two lines: "夜读" on the first line and "NIGHT READING" on the second. Both lines must be correctly spelled, sharp, evenly spaced, and easily readable. Frame the sign large enough to read, in the near left third; the window and readers fill the rest. Calm amber interior versus blue exterior, believable glass reflections that do not obscure people or lettering. No other legible text, shop branding, banners, posters or sales context. No advertisement, brand, logo, fox, product display, promotional graphics, or watermark.

REST API

Quick Start

This flow submits one valid Grok Imagine Image Quality Text to Image request, captures its task_id, and reads image URLs after the asynchronous task ends.

Connect to the Vidgo API

Create an API key, keep it on the server, and send it as Authorization: Bearer VIDGO_API_KEY.

Endpoint: POSThttps://api.vidgo.ai/api/generate/submit
Authentication: Authorization: Bearer VIDGO_API_KEY

Submit one image task

Send the current model ID, prompt to the unified submit endpoint. The example keeps only first-call fields.

Submit request

curl --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "xai/grok-imagine-image-quality/text-to-image",
  "input": {
    "prompt": "A ceramic cup on a wooden table.",
    "aspect_ratio": "1:1",
    "resolution": "1K",
    "n": 1
  }
}'

Wait for the image result

Read data.task_id and poll only while status is not_started or running. Stop on both finished and failed.

Track status

GET https://api.vidgo.ai/api/generate/status/{task_id}

Poll every 2–5 seconds and back off on 429/5xx. Or provide top-level callback_url to receive a terminal-state callback.

not_startedrunningfinishedfailed

Submit response

{
  "code": 200,
  "data": {
    "task_id": "example-task-id",
    "status": "running",
    "created_time": "2026-09-24T10:00:00Z"
  }
}

Finished response

{
  "code": 200,
  "data": {
    "task_id": "example-task-id",
    "status": "finished",
    "created_time": "2026-09-24T10:00:00Z",
    "files": [
      {
        "file_url": "https://example.com/output.png",
        "file_type": "image"
      }
    ]
  }
}

Complete runnable polling example

Expand for the same payload end to end, including business-code, missing-task-ID, failed-state, unexpected-state, and network-error handling.

Complete runnable polling example

set -euo pipefail
: "${VIDGO_API_KEY:?Set VIDGO_API_KEY in your environment}"

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "xai/grok-imagine-image-quality/text-to-image",
  "input": {
    "prompt": "A ceramic cup on a wooden table.",
    "aspect_ratio": "1:1",
    "resolution": "1K",
    "n": 1
  }
}
JSON
)

SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

TASK_ID=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.data.task_id // .task_id // empty')
if [ -z "$TASK_ID" ]; then
  printf 'Submit response did not include task_id:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi

while true; do
  STATUS_RESPONSE=$(curl --silent --show-error --fail-with-body \
    --url "https://api.vidgo.ai/api/generate/status/$TASK_ID" \
    --header "Authorization: Bearer $VIDGO_API_KEY")
  STATUS=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.data.status // .status // empty')

  case "$STATUS" in
    finished)
      printf '%s' "$STATUS_RESPONSE" | jq -r '(.data.files // .files // [])[]?.file_url'
      break
      ;;
    failed)
      printf '%s' "$STATUS_RESPONSE" | jq -r '.data.error_message // .error_message // "Generation failed"' >&2
      exit 1
      ;;
    not_started|running)
      sleep 2
      ;;
    *)
      printf 'Unexpected task status: %s
' "$STATUS" >&2
      exit 1
      ;;
  esac
done

Request Parameters

This table lists only public fields inside input. See Quick Start for top-level model and optional callback_url usage.

Field	Type	Required	Default	Description
input.prompt	string	Yes	–	Required non-blank string. No length maximum is documented upstream.
input.aspect_ratio	string	No	1:1	Supports 1:1, 2:3, 3:2, 9:16, 16:9. Default: 1:1.
input.resolution	string	No	1K	Choose 1K or 2K. Default: 1K.
input.n	integer	No	1	Output image count, integer at least 1, default 1. No maximum is documented upstream.

Response Fields

The submit response identifies the task. The status response supplies image files or failure details at a terminal state.

Field	Type	Description
code	integer	Business response code; 0 or 200 indicates success.
message	string	Optional human-readable response message.
data.task_id	string	Unique task ID used for status queries after submission.
data.status	string	not_started, running, finished, or failed.
data.created_time	string	Task creation timestamp.
data.progress	integer	Progress from 0–100 in status responses.
data.files[]	array	Generated files in returned order when finished.
data.files[].file_url	string	Direct generated-image URL.
data.files[].file_type	string	Image results are identified as image.
data.files[].watermark_url	string \| null	Optional watermarked-file URL when present.
data.error_message	string \| null	Readable failure reason for failed tasks.

Task Lifecycle

Only the first two states may be polled again. finished and failed are terminal.

not_started
Accepted and queued; wait before checking again.
running
Image generation is in progress; continue polling.
finished
Successful terminal state; validate files and read image URLs.
failed
Failure terminal state; read error_message and stop.

Polling and Errors

Authentication401 means the key is missing or invalid. Fix authentication before resubmitting and never expose the key in browser code.
Validation400 indicates an invalid model ID, field, enum, length, or image count.
Polling intervalStart at 2–5 seconds, use exponential backoff for 429/5xx, and enforce an overall timeout.
Terminal statesStop after finished or failed. Treat finished without a valid image as a separate error.
Callback optionProvide a public top-level callback_url to receive a flat task object after finished or failed.
Network errorsDistinguish request timeout, network failure, and task failure. Retry the same query only when safe.

Model Specifications

Specification	Value	Details
Input mode	Prompt only	The prompt is the complete creative source.
Output	Image	Returns one or more image files asynchronously.
Resolution	1K · 2K	Defaults to 1K.
Aspect ratio	5 ratios	Supports 1:1, 2:3, 3:2, 9:16, 16:9.
N	≥ 1	Integer output count, default 1. No maximum is documented upstream.
Billing	Total credits = output count × resolution rate + reference image count × 2. Reference fees apply only to Edit, once per request, without multiplying by output count. 1 credit = $0.005.	1 credit = $0.005; the backend remains authoritative for final billing.

Grok Imagine Image Quality Text to Image API documentation

Grok Imagine Image Quality Text to Image Overview

Grok Imagine Image Quality Text to Image is xAI's high-fidelity image generation model designed for photorealistic visual synthesis. It excels in authentic photographic realism, precise prompt adherence, crisp multilingual typography rendering, and refined optical texture, offering 1K and 2K resolution tiers with 5 versatile aspect ratios (1:1, 2:3, 3:2, 9:16, and 16:9).

Why Choose Grok Imagine Image Quality Text to Image?

Photorealistic Fidelity and Nuanced LightingTuned specifically for high fidelity, capturing authentic skin textures, natural imperfections, and intricate lighting interactions that bring photographic realism to every generation.
Sharp Multilingual Typography RenderingOvercomes traditional image generation limitations by rendering crisp, legible text and short titles across multiple languages for advertising and graphic design.
Optically Grounded Prompt AdherenceDeeply interprets camera focal lengths, aperture depths of field, complex studio lighting setups, and spatial perspectives for accurate creative translation.
1K and 2K Resolutions with Multiple RatiosProvides 1K for rapid conceptual drafts and 2K for sharp final assets, alongside native support for 1:1, 2:3, 3:2, 9:16, and 16:9 aspect ratios.
Transparent Metered Pricing with Automatic RefundsCharges credits precisely based on output resolution and quantity, with automatic full refunds if a task encounters an upstream error.

Parameters

Parameter	Requirement	Description
prompt	Required	Required non-blank string. No length maximum is documented upstream.
aspect_ratio	Optional	Supports 1:1, 2:3, 3:2, 9:16, 16:9. Default: 1:1. Default`1:12:33:29:1616:9`
resolution	Optional	Choose 1K or 2K. Default: 1K. Default`1K2K`
n	Optional	Output image count, integer at least 1, default 1. No maximum is documented upstream. Default`1`

How to Use

Formulate Structured PromptsDescribe your scene using a structured hierarchy covering subject details, lighting environment, camera specifications, and typography in quotation marks.
Select Output ResolutionChoose between 1K for faster prototyping and 2K for production-grade sharpness and fine microscopic detail.
Configure Aspect Ratio and Image CountSet the aspect ratio (1:1, 2:3, 3:2, 9:16, or 16:9) to fit your destination platform, and specify n for the number of image variations.
Submit Asynchronous Generation TaskSend your request to obtain a unique task_id, processed asynchronously across dedicated high-fidelity rendering clusters.
Poll and Retrieve Final ImagesPoll the task status endpoint until finished, then access high-resolution image URLs for immediate download.

Pricing

Total credits = output count × resolution rate + reference image count × 2. Reference fees apply only to Edit, once per request, without multiplying by output count. 1 credit = $0.005.

Usage	Rate	Details
1K output	8 credits / output image	$0.040 / image · Official $0.050 · 20%
2K output	11 credits / output image	$0.055 / image · Official $0.070 · 21%

Best Use Cases

Commercial and Food PhotographySimulate studio lighting and shallow depth of field to produce mouth-watering culinary imagery and high-end commercial product visuals.
Marketing Posters and Typography GraphicsUtilize exceptional typography rendering to generate complete marketing banners, packaging mockups, and promotional graphics with readable text.
Cinematic Concept Art and StoryboardsTranslate narrative scripts and worldbuilding concepts into evocative, film-grade production stills with believable mood and composition.
Social Media and Multi-Platform VisualsEasily produce matching assets across 9:16 vertical reels, 16:9 widescreen banners, and 1:1 square feed graphics from a unified prompt style.

Pro Tips

Layering specific optical specs (e.g., "shot on 85mm lens, f/1.4, shallow depth of field, warm morning rim light") produces significantly more cinematic perspective and authentic depth.
To render readable typography, enclose exact words in half-width double quotation marks (e.g., "ARTISAN ROAST") and specify the intended sign, packaging, or layout context.
For authentic photographic texture, include "candid style, subtle film grain, natural skin texture and imperfections" to avoid plastic, over-smoothed digital artifacts.
Generate multiple variations in a single request using the n parameter to explore composition variations under identical lighting and stylistic conditions.

Notes

The prompt parameter is required, non-blank, and supports comprehensive optical, compositional, and stylistic descriptions.
Output options support 1K (default) and 2K resolutions, alongside five aspect ratios: 1:1 (default), 2:3, 3:2, 9:16, and 16:9.
Processing is asynchronous; check status using the returned task_id via polling or configure a callback_url for webhook delivery.

Related Models

Qwen Image 3.0 Text-to-ImageAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Qwen Image 3.0 Image EditAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Qwen Image 3.0 Pro Text-to-Image preview

Qwen Image 3.0 Pro Text-to-ImageAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Qwen Image 3.0 Pro Image EditAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Seedream 5.0 Pro Text-to-ImageThe ByteDance Seedream 5.0 Pro API generates 1K or 2K JPEG/PNG images from prompts and edits one to ten ordered HTTP(S) image inputs.

Seedream 5.0 Pro Image EditThe ByteDance Seedream 5.0 Pro API generates 1K or 2K JPEG/PNG images from prompts and edits one to ten ordered HTTP(S) image inputs.

GPT Image 2 Text-to-ImageOpenAI image generation and editing with quality tiers, flexible size controls, and single-image outputs up to 4K.

GPT Image 2 Image EditOpenAI image generation and editing with quality tiers, flexible size controls, and single-image outputs up to 4K.

Nano Banana 2 Text-to-ImageGoogle's next-gen image model powered by Gemini 3.1 Flash with native 2K/4K resolution, chain-of-thought reasoning, precise multi-language text rendering, and up to 14 reference images.

Nano Banana 2 Image EditGoogle's next-gen image model powered by Gemini 3.1 Flash with native 2K/4K resolution, chain-of-thought reasoning, precise multi-language text rendering, and up to 14 reference images.

Nano Banana Pro Text-to-ImageGoogle's Nano Banana Pro image model powered by Gemini 3 Pro for 4K generation, strong text rendering, multi-image blending, and production-ready image editing.

Nano Banana Pro Image EditGoogle's Nano Banana Pro image model powered by Gemini 3 Pro for 4K generation, strong text rendering, multi-image blending, and production-ready image editing.

GPT Image 1.5OpenAI's latest image model with 4x speed, precision editing, and superior text rendering.

Z-ImageAlibaba's efficient 6B-parameter image model with sub-second generation and exceptional Chinese-English bilingual text rendering capabilities.

Seedream 4.5ByteDance's unified 4K image generation and editing model with professional-grade text rendering and commercial photography quality.

Seedream 5.0 LiteByteDance Seedream 5.0 Lite family on Vidgo API: seedream-5.0-lite for text-to-image and seedream-5.0-lite-edit for image-to-image editing, with 2K/3K presets, custom sizes, and up to 10 reference images.

Grok Imagine Image Quality Text to Image API frequently asked questions

What is the Grok Imagine Image Quality Text to Image API?

Grok Imagine Image Quality Text to Image is an xAI model for high-fidelity image generation from text prompts. It transforms text prompts into photorealistic images at 1K and 2K resolutions, featuring superior prompt adherence, nuanced optical lighting control, sharp multilingual text rendering, and support for 5 standard aspect ratios. Built on xAI's quality-focused multimodal generative architecture, it adheres strictly to technical camera and lighting instructions while preserving natural textures, subtle imperfections, and cohesive visual composition. You can call it programmatically or try it from the playground above.

Can Grok Imagine Image Quality Text to Image render legible text in images?

Yes. Text rendering is one of this quality tier's standout strengths. When generating packaging, marketing banners, and street signage, it delivers crisp, legible text across multiple languages. Enclosing target wording in half-width double quotation marks (such as "SUMMER SALE") and describing its placement will yield the most accurate typographic layout.

How should I choose between 1K and 2K resolution in Grok Imagine Image Quality Text to Image?

The 1K resolution tier processes faster at a lower credit cost (8 credits per output image), making it optimal for rapid creative prototyping, social media drafts, or high-volume generation. The 2K tier (11 credits per output image) delivers enhanced sharpness, fine texture rendition, and edge clarity, ideal for print deliverables, hero marketing banners, and finished visual assets.

How can I prompt Grok Imagine Image Quality Text to Image for maximum photorealism?

Opt for concrete, optically grounded descriptions rather than generic quality adjectives. Specify camera focal lengths (e.g., 35mm or 85mm), lens aperture (e.g., f/1.4 or f/2.8), lighting angles, and directional falloff. Adding terms like "candid photography" and "natural skin imperfections" helps guide the model toward realistic, lifelike outputs.

Which aspect ratios does Grok Imagine Image Quality Text to Image support?

The model supports five native aspect ratios: 1:1 (square, default), 2:3 (vertical portrait), 3:2 (classic landscape photography), 9:16 (mobile vertical video and stories), and 16:9 (widescreen cinematic framing). Set the aspect_ratio parameter during submission, and the model adjusts composition and spatial depth accordingly.

Which model successor should I migrate to from Grok Imagine Image Quality Text to Image?

According to xAI's lifecycle roadmap, Grok Imagine Image Quality is scheduled to complete its transition on November 2, 2026. Users and developers are encouraged to migrate workflows to Grok Imagine Image 2.0, which offers further advancements in graphic typography rendering, detailed prompt following, and generational efficiency.

Can outputs from Grok Imagine Image Quality Text to Image be used commercially?

Yes. Under xAI's licensing terms, users retain full ownership and usage rights over generated output images, enabling commercial use across advertising, digital products, marketing collateral, and merchandise. Users remain responsible for prompt compliance and third-party intellectual property guidelines.