Wan 2.2 Text To Image Realism

Wan 2.2 Text To Image Realism

Playground

WAN 2.2 delivers ultra-realistic text-to-image generation, converting prompts into photoreal images with high fidelity and detail. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Generate photorealistic images from detailed text descriptions with Wan 2.2 Realism. This specialized model excels at creating lifelike scenes, authentic human subjects, and natural environments — perfect for when you need images that look like real photographs.

Why It Looks Great

• Photorealistic focus: Optimized specifically for realistic, photograph-like outputs.
• Detailed human rendering: Excels at natural skin tones, expressions, and group compositions.
• Custom dimensions: Precise control over width and height for any aspect ratio.
• High resolution support: Generate images up to 1280×720 and beyond.
• Prompt Enhancer: Built-in tool to refine and expand your descriptions automatically.
• Reproducible results: Use the seed parameter to recreate exact outputs or explore variations.

Parameters

How to Use

1. Write your prompt — describe the scene in detail, including people, setting, lighting, and atmosphere.
2. Use Prompt Enhancer (optional) — click to automatically enrich your description.
3. Set dimensions — adjust width and height sliders to your desired resolution.
4. Set seed (optional) — use -1 for random, or a specific number to reproduce results.
5. Choose output format — select jpeg for smaller files or png for higher quality.
6. Run — click the button to generate.
7. Download — preview and save your realistic image.

Pricing

Flat rate per image generation.

Examples

Best Use Cases

• Lifestyle & Stock Photography — Generate authentic-looking lifestyle scenes and stock imagery.
• Group Portraits — Create realistic multi-person compositions with natural interactions.
• Environmental Scenes — Produce believable outdoor settings, gatherings, and events.
• Marketing & Advertising — Generate photorealistic visuals for campaigns without photoshoots.
• Concept Visualization — Visualize realistic scenarios for presentations and pitches.

Example Prompts

• “A group of four women are seated around a wooden picnic table outdoors at a backyard gathering. The woman in the foreground, light-skinned and young adult, has shoulder-length light brown hair and a friendly, smiling expression. She’s wearing a white, sleeveless top.”
• “Professional headshot of a middle-aged businessman in a navy suit, soft studio lighting, neutral gray background, confident expression”
• “Family enjoying breakfast in a sunny modern kitchen, natural morning light through windows, warm and authentic atmosphere”
• “Two colleagues having a conversation in a contemporary office space, natural poses, professional but relaxed mood”
• “Street photographer capturing city life, candid moment, golden hour lighting, urban background with bokeh”

Pro Tips for Best Results

• Be extremely detailed — describe physical features, clothing, expressions, and positioning.
• Include lighting details — “natural sunlight”, “soft studio lighting”, “golden hour”.
• Specify skin tones, ages, and distinguishing features for accurate human rendering.
• Describe the environment and background to ground the scene in reality.
• Use landscape dimensions (1280×720) for group scenes, portrait (720×1280) for individual shots.
• The more specific your prompt, the more realistic and controlled the output.

Notes

• This model is optimized for realism — for artistic or stylized outputs, consider other models.
• Higher detail in prompts generally produces more accurate and realistic results.
• Generation time may vary based on resolution and current queue load.
• For multi-person scenes, describe each person’s position and appearance clearly.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "size": "1024*1024",
  "seed": -1,
  "output_format": "jpeg"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.2/text-to-image-realism" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters