Docs / Image Generation

Image Generation

Generate images from text prompts through the OpenAI-compatible POST /v1/images/generations endpoint. The request returns image URLs directly; no polling required.

#Models

Model ID	Family	Notes
grok-imagine-image	Grok Imagine	Fast image generation
grok-imagine-image-quality	Grok Imagine	Higher quality tier

#API usage

cURL

terminalbash

curl https://relaydance.com/v1/images/generations \
  -H "Authorization: Bearer $RELAYDANCE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-imagine-image",
    "prompt": "A serene Japanese garden with cherry blossoms, watercolor style",
    "n": 1
  }'

Python

image.pypython

import requests, os

response = requests.post(
    "https://relaydance.com/v1/images/generations",
    headers={
        "Authorization": f"Bearer {os.environ['RELAYDANCE_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
        "model": "grok-imagine-image",
        "prompt": "A serene Japanese garden with cherry blossoms, watercolor style",
        "n": 1,
    },
)
data = response.json()
print(data["data"][0]["url"])

OpenAI SDK

image.tstypescript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.RELAYDANCE_API_KEY,
  baseURL: "https://relaydance.com/v1",
});

const res = await client.images.generate({
  model: "grok-imagine-image",
  prompt: "A futuristic city at dusk, cinematic lighting",
  n: 1,
});
console.log(res.data[0].url);

#Response

response.jsonjson

{
  "data": [
    { "url": "https://..." }
  ]
}

#Request parameters

Parameter	Type	Description
`model`	string	Image model ID
`prompt`	string	Text description of the image to generate
`n`	integer	Number of images to generate. Default: 1
`size`	string	Optional output size, model-dependent (e.g. "1024x1024")

#Tips for better results

Be specific: describe style, composition, lighting and mood in your prompt
Name a style: keywords like "photorealistic", "watercolor", "anime" steer the output strongly
Iterate: small prompt changes can shift results a lot; generate, adjust, repeat

Generated images make good reference inputs for video: pass the URL in metadata.content[] of a video task. See Image-to-Video.

←Video Generation LLM / Chat→