P H3 Image Image Edit API Documentation

P H3 Image Image Edit API Documentation

Playground

P-H3-Image by Pruna AI, built on the MiniMax H3 open weights, edits an image from a text instruction with up to 8 extra reference images, keeping faces, products and details consistent, at 1K to 4K resolution. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.

Features

P-H3-Image is Pruna AI’s image generation and editing model built on the MiniMax H3 open weights. This endpoint edits the first input image from a text instruction, using up to 8 more images as references, and renders the result from 1K up to 4K.

Powered by MiniMax H3.

Why Choose This?

• Photorealistic by default
  Natural skin, light and materials, with strong prompt adherence for documentary, editorial and cinematic looks.

• Identity and product consistency
  Faces, hairstyles, outfits and product details carry over from the input images.

• Up to 4K output
  Choose 1k, 1.5k, 2k, 3k or 4k; 3k sits closest to the model’s sweet spot for fine detail.

• Built-in prompt enhancement
  Short requests are expanded into detailed prompts automatically, so a one-line idea still renders well.

• Multi-image references
  Combine up to 9 images, addressed in the prompt as <Picture 1>, <Picture 2>, and so on.

Parameters

How to Use

1. Upload images — The image to edit first; add references for outfits, products or more people.
2. Describe the result — Say what should change and what should stay, e.g. Put her in a sunflower field at golden hour wearing a wide straw hat; keep her face and braided hair.
3. Choose resolution — 1k–2k for iteration, 3k or 4k for final assets.
4. Submit — Receive the edited image URL.

Pricing

Pricing is based on the selected resolution.

The prices below include the first input image. Each additional input image costs $0.005, regardless of resolution.

Pro Tips

• Describe the scene concretely (subject, setting, light, lens); avoid hype words, which tend to make images worse.
• Name what each reference is for, e.g. the hat from <Picture 2>.
• Short quoted text such as a sign or a title renders well; long or small text is less reliable.
• Use 3k or 4k for hands, small background figures and fine textures.
• Fix the seed to iterate on wording without changing the composition.

Related Models

• P-H3-Image Text-to-Image
• P-H3-Image Text-to-Image LoRA
• P-H3-Image Edit LoRA
• MiniMax H3 Text-to-Video — Video generation with native audio on the same model family.

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",
  "images": [
    "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
  ],
  "aspect_ratio": "1:1",
  "resolution": "1k",
  "output_format": "jpeg"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/p-h3-image/image-edit" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters