Bytedance Seed3d 2.0 Image To 3d API Documentation

Bytedance Seed3d 2.0 Image To 3d API Documentation

Playground

ByteDance Seed3D 2.0 is a fast AI image-to-3D generation model that converts a single reference image into a GLB 3D asset. Ready-to-use REST inference API for 3D asset generation, product visualization, game assets, AR/VR content, digital twins, creative design, and professional image-to-3D workflows with simple integration, no coldstarts, and affordable pricing.

Features

ByteDance Seed3D 2.0 Image-to-3D converts a single reference image into a downloadable GLB 3D asset. Upload an image of the object, choose a mesh subdivision level, and receive a WaveSpeed CDN output URL.

Why Choose This?

• Single-image 3D generation Convert one reference image into a downloadable GLB 3D asset.

• Subdivision level control Choose low, medium, or high subdivision level depending on the desired mesh detail.

• GLB output The model returns a GLB file suitable for common 3D workflows.

Parameters

How to Use

1. Upload your image — Provide a reference image URL for the object you want to reconstruct.
2. Choose subdivision level (optional) — Select low, medium, or high. The default is medium.
3. Submit — Generate the GLB 3D asset.
4. Download the result — The generated GLB URL is returned in the standard WaveSpeed prediction response outputs array.

Pricing

Pricing is fixed at $0.40 per generation.

Billing Rules

• Each request generates one GLB 3D asset.
• Each generation costs $0.40.
• The price is fixed and does not change based on subdivision_level.

Best Use Cases

• Image-to-3D asset creation — Generate a 3D asset from a single object image.
• Product 3D previews — Convert product images into GLB assets for 3D display workflows.
• Prototype asset generation — Quickly create draft 3D assets from reference images.
• Object reconstruction — Reconstruct one main object from a clean, centered input image.
• GLB-based workflows — Generate assets in GLB format for downstream 3D use.

Pro Tips

• Use clean, centered images with one main object.
• Avoid cluttered backgrounds when possible.
• Make sure the object is clearly visible in the input image.
• Use the default medium subdivision level unless you specifically need lower or higher mesh subdivision.
• Ensure the image URL is publicly accessible.

Notes

• image is the only required field.
• subdivision_level is optional.
• Default subdivision_level is medium.
• Output is returned as a GLB file.
• Ensure uploaded image URLs are publicly accessible.

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",
  "subdivision_level": "medium"
}
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/bytedance/seed3d-2.0/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

Response Parameters

Result Request Parameters

Result Response Parameters