Alibaba Wan 2.6 Text To Image

Alibaba Wan 2.6 Text To Image

Playground

WAN 2.6 Text-to-Image generates high-quality images from natural-language prompts with strong prompt adherence and clean composition. It supports multiple aspect ratios and size control, seed-based reproducibility, and flexible styles (photorealistic to illustrative) for ads, product shots, and social visuals. Built for stable production use with a ready-to-use REST API, no cold starts, and predictable pricing.

Features

Wan 2.6 Text-to-Image (/wan-2.6/text-to-image) is ’s text-to-image generation model for creating high-quality visuals from a single natural-language prompt. It’s built for practical creative workflows—concept art, product visuals, portraits, and stylized imagery—where you want strong prompt adherence plus flexible custom sizing.

Why it stands out

• Fast, one-shot text-to-image generation Generate an image in a single run for quick ideation and production workflows.

• Custom width × height output Set width and height directly (within the endpoint’s limits) to match banners, thumbnails, posters, or social formats.

• Prompt expansion for better results Enable prompt expansion to automatically enrich short prompts with useful detail for more coherent compositions.

• Seeded iteration Use a fixed seed to refine style and layout with more repeatable variations.

Parameters

How to use

1. Write a clear prompt (subject + setting + style).
2. Choose width and height that match your target aspect ratio.
3. Turn on enable_prompt_expansion if your prompt is short or under-specified.
4. Set a seed if you want repeatable iterations (keep the same seed while you tweak the prompt).
5. Click Run, review the result, and iterate.

Prompt tips

• Start with subject + environment + style: “A modern tea shop interior, warm afternoon light, minimalist wood design, cinematic photography.”
• Add camera / composition when framing matters: “wide shot, shallow depth of field, 35mm film look.”
• Keep instructions positive and specific (what you want to see, not what you fear).

Pricing

• $0.03 per generated image

Notes

• Output sizing is limited by the endpoint’s current constraints (for example, width/height bounds and aspect-ratio limits). If a size fails, reduce resolution or choose a more standard aspect ratio.
• Enabling prompt expansion can improve quality for short prompts, but may add a little latency.
• Returned image URLs may be time-limited—save outputs if you need long-term storage.

Related Models

• Wan 2.5 Text-to-Image — A proven Wan text-to-image model for reliable, cost-stable AI image generation with a similar prompt-first workflow.
• Seedream V4 Text-to-Image — A style-consistent text-to-image generator for posters, campaigns, and high-volume brand-friendly illustration batches.
• FLUX.2 Turbo Edit — A fast natural-language image editing model for precise image-to-image transformations, brand color control, and iterative creative revisions.
• Google Nano Banana Pro Edit — High-fidelity prompt-based image editing for composition-preserving changes, product visuals, and reliable on-image text handling.

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",
  "enable_prompt_expansion": false,
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/alibaba/wan-2.6/text-to-image" \
  -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