Minimax H3 Image Text To Image API Documentation

Minimax H3 Image Text To Image API Documentation

Playground

MiniMax H3 Image, the open-weights release of MiniMax’s image model, generates detailed, photorealistic images from text prompts, native output from 1K to 4K. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

MiniMax H3 Image Open-Weights Text-to-Image generates images from text prompts using the open-weights release of MiniMax H3 Image. It supports multiple aspect ratios, resolutions from 1k to 4k, and several output formats.

Why Choose This?

• Open-weights model — Run the open-weights release of MiniMax H3 Image.
• Up to 4K output — Choose 1k, 1.5k, 2k, 3k, or 4k.
• Flexible framing — Generate square, portrait, landscape, tall, wide, and ultrawide compositions.
• Multiple output formats — Export generated images as JPEG, PNG, or WebP.
• Seed control — Use a fixed seed for more reproducible generation.

Parameters

How to Use

1. Write a prompt — Describe the subject, setting, composition, lighting, and style.
2. Choose an aspect ratio — Select the framing that fits the intended output.
3. Choose a resolution — Select 1k, 1.5k, 2k, 3k, or 4k.
4. Choose an output format optional — Select JPEG, PNG, or WebP.
5. Set a seed optional — Use a fixed seed when comparing prompt or parameter changes.
6. Submit — Generate and retrieve the image.

Pricing

Pricing is based on the selected resolution.

Aspect ratio, output format, and seed do not add separate charges.

Best Use Cases

• General image generation — Create images from text descriptions.
• Advertising concepts — Generate visuals for marketing and promotional workflows.
• Product concepts — Create product-focused images and presentation concepts.
• Editorial imagery — Generate visuals for publishing and content workflows.
• High-resolution output — Use higher resolution tiers when larger output sizes are required.

Pro Tips

• Describe the main subject and composition clearly.
• Include lighting, environment, materials, and style when they are important to the result.
• Use lower resolutions for iteration and higher resolutions when a larger output is required.
• Use a fixed seed when comparing prompt changes.
• Choose the aspect ratio based on the intended layout or publishing format.

Notes

• prompt is required.
• aspect_ratio defaults to 1:1.
• resolution defaults to 1k.
• Supported resolutions are 1k, 1.5k, 2k, 3k, and 4k.
• output_format defaults to jpeg.
• Use seed=-1 for random seed selection.
• Higher resolution tiers use higher pricing.

Related Models

• MiniMax H3 Image Open-Weights Edit — Edit existing images using MiniMax H3 Image open weights.
• MiniMax H3 Image Open-Weights Text-to-Image LoRA — Generate images with custom LoRA weights.
• MiniMax H3 Image Open-Weights Edit LoRA — Edit images with custom LoRA weights.
• MiniMax H3 Image — Generate images through the MiniMax H3 Image API.
• P-H3-Image — Generate images with the P-H3-Image implementation based on MiniMax H3 open weights.

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",
  "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/minimax-h3-image/text-to-image" \
  -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