How to Generate an Image

How to Generate an Image

Step-by-step guide to generating images with WaveSpeedAI.

Quick Start

Web Interface

  1. Go to wavespeed.ai/models
  2. Select a text-to-image model (e.g., FLUX, Seedream, Nano Banana)
  3. Enter your prompt
  4. Click Run

API

curl --location --request POST 'https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-dev' \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
  "prompt": "A majestic mountain landscape at sunset, cinematic lighting",
  "size": "1024*1024",
  "num_inference_steps": 28,
  "guidance_scale": 3.5,
  "num_images": 1,
  "seed": -1
}'
ModelBest ForSpeed
FLUX DevGeneral purpose, high qualityFast
FLUX SchnellQuick iterationsVery fast
SeedreamVersatile, good prompt followingMedium
Nano Banana ProHigh quality, realisticFast

Common Parameters

ParameterDescriptionExample
promptWhat to generate”A cat in space”
widthOutput width1024
heightOutput height768
num_inference_stepsQuality (more = better)20-50
seedFor reproducibility12345

Example: High Quality Image

{
  "prompt": "A fluffy orange cat sitting on a windowsill, soft afternoon sunlight streaming through the window, bokeh background, photography style, 4K, high detail",
  "width": 1024,
  "height": 1024,
  "num_inference_steps": 30
}

Example: Multiple Variations

Generate multiple images with different seeds by making separate API calls:

# First variation (seed: 1)
curl --location --request POST 'https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-dev' \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
  "prompt": "A cat in space",
  "size": "1024*1024",
  "num_inference_steps": 28,
  "guidance_scale": 3.5,
  "seed": 1
}'
 
# Second variation (seed: 2)
curl --location --request POST 'https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-dev' \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
  "prompt": "A cat in space",
  "size": "1024*1024",
  "num_inference_steps": 28,
  "guidance_scale": 3.5,
  "seed": 2
}'

Tips for Better Results

  1. Be specific — Include details about subject, style, lighting
  2. Use Prompt Enhancer — Click the button to optimize your prompt
  3. Start small — Test with lower resolution first
  4. Check model README — Each model has specific tips

See How to Write Better Prompts for more tips.

© 2025 WaveSpeedAI. All rights reserved.