Tripo3d H3.0 Image To 3d API Documentation

Tripo3d H3.0 Image To 3d API Documentation

Playground

Tripo3D H3.0 Image-to-3D turns a single image into a 3D model with Tripo’s previous-generation H3.0 model (up to 2 million faces), kept for consistency with earlier assets. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Tripo3D H3.0 Image-to-3D generates a 3D model from a single reference image. Use a clear image of the object, then configure geometry, textures, topology, and optional PBR materials for the final asset.

Why Choose This?

• Image-to-3D generation — Create a 3D model from a single reference image.
• Texture and PBR options — Generate untextured geometry, texture maps, or PBR materials.
• Geometry control — Choose standard or detailed geometry and set optional face limits.
• Topology options — Generate triangle meshes, quad meshes, or smart low-poly assets.
• Editable parts — Generate separate untextured parts for downstream editing.
• Texture controls — Choose texture quality, alignment, UV generation, and optional lighting cleanup.

Parameters

How to Use

1. Upload an image — Choose a clear view of the object with minimal obstruction.
2. Choose materials — Keep textures and PBR enabled, or disable both for an untextured model.
3. Choose quality — Select texture and geometry quality based on the level of detail you need.
4. Configure topology optional — Use quad, smart low-poly, face limits, or separate parts when needed.
5. Submit — Generate and retrieve the completed 3D model.

Pricing

Pricing is per generated model and depends on the selected options.

Base Price

Optional Add-ons

The default configuration costs $0.36 per model.

PBR does not add a separate charge beyond the textured model price.

detailed and extreme texture charges apply only when textures are generated.

When generate_parts=true, the model uses the untextured base price and the separate-parts add-on. Texture and quad charges do not apply.

Best Use Cases

• Product reconstruction — Turn product photos into reusable 3D assets.
• Game assets — Generate textured, low-poly, or quad-based models for game workflows.
• E-commerce 3D — Create 3D product assets from existing photography.
• Rendering assets — Generate models with textures and optional PBR materials.
• Editable models — Create separated parts for additional modeling and refinement.
• High-detail assets — Combine detailed geometry with higher-quality textures.

Pro Tips

• Use a clear image with the object fully visible and minimal obstruction.
• A clean background can help the model identify the object more reliably.
• Use detailed geometry only when additional mesh detail is needed.
• Use fast textures for quicker iteration and extreme when 8K texture detail is required.
• Use smart_low_poly for simpler objects that need lightweight topology.
• Use generate_parts when individual components need to remain separately editable.

Notes

• Supported input formats are PNG, JPEG, and WebP.
• Maximum input image size is 20 MB.
• Disable both texture and pbr for an untextured model.
• Enabling pbr also enables texture generation.
• fast texture quality requires texture_version=v3.5-20260815.
• delight applies only to texture version v3.5-20260815.
• quad=true generates an FBX quad mesh and defaults to approximately 10,000 faces when face_limit is not specified.
• generate_parts requires texture=false and pbr=false.
• Do not combine generate_parts with quad or smart_low_poly.
• Separate parts are generated as untextured triangle meshes.
• auto_size estimates real-world scale and outputs the model size in meters.
• orientation=align_image applies only to textured models.
• Face limits depend on the selected geometry and topology settings.

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'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "texture": true,
  "pbr": true,
  "texture_quality": "standard",
  "texture_version": "v3.0-20250812",
  "delight": true,
  "geometry_quality": "standard",
  "quad": false,
  "smart_low_poly": false,
  "generate_parts": false,
  "texture_alignment": "original_image",
  "orientation": "default",
  "enable_image_autofix": false,
  "auto_size": false,
  "export_uv": true
}
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/tripo3d/h3.0/image-to-3d" \
  -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