Openai Gpt 6 Astra Image To 3d API Documentation

Openai Gpt 6 Astra Image To 3d API Documentation

Playground

GPT-6 Astra Image-to-3D generates textured 3D models or scenes from reference images, with optional text guidance for image-guided 3D asset creation, product visualization, concept design, virtual scenes, and production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

GPT-6 Astra Image to 3D turns reference images into a textured 3D object or scene that can be rotated, explored, and developed further. Provide one or more images, add optional prompt guidance, set the subject scale, and generate a portable GLB model for downstream 3D workflows.

Why Choose This?

• Image-to-3D generation
  Convert reference images into a generated 3D object or scene.

• Object and scene creation
  Generate standalone products, props, vehicles, interiors, buildings, environments, or broader spatial concepts.

• Material-aware output
  Guide materials such as timber, stone, glass, painted metal, fabric, ceramic, or worn surfaces through the prompt.

• PBR texture support
  The exported model includes geometry and embedded PBR textures.

• Portable GLB output
  Download a GLB model that can be viewed in compatible browser viewers or imported into 3D applications.

Parameters

How to Use

1. Provide reference images — Upload images of the subject or scene you want to convert into 3D.
2. Add prompt guidance optional — Describe the subject, materials, hidden details, or surrounding environment.
3. Set size optional — Define the longest subject dimension when a specific physical scale is needed.
4. Submit — Generate the 3D result.
5. Download the GLB — Retrieve the generated GLB model URL from outputs.

Pricing

Pricing is fixed at $8 per generation.

Output

Each successful request returns a GLB model URL in outputs.

The generated model includes geometry and embedded PBR textures. Tasks have a 120-minute timeout.

Best Use Cases

• Product visualization — Turn product references into 3D assets for inspection and presentation.
• Concept design — Explore objects, vehicles, architecture, props, interiors, and environments in 3D.
• Scene prototyping — Generate spatial concepts such as rooms, courtyards, workshops, hangars, or landscapes.
• Interactive experiences — Create starting assets for viewers, prototypes, games, or immersive workflows.
• Creative iteration — Compare multiple design directions from the same set of reference images.

Pro Tips

• Use clear, consistent reference images of the same subject.
• Add multiple angles when shape and proportions matter.
• Use the prompt to explain details that are hidden, ambiguous, or not visible in the images.
• For scenes, describe both the main subject and the surrounding space.
• Include material details such as surface wear, glass, metal, stone, fabric, or wood when appearance matters.
• Inspect the first result from several angles before refining the next prompt.
• Generated assets may need cleanup for production use.
• The result is a generated interpretation, not a dimensionally exact scan.

Example Prompt

A compact exploration spacecraft resting on three landing legs inside a service hangar. Give it a rounded ceramic hull, a smoked-glass cockpit, recessed engine nozzles, and visible panel seams. Include maintenance carts and a grated floor, with enough space to inspect the craft from all sides.

Notes

• images is required.
• Up to 10 reference images are supported.
• prompt is optional and supports up to 4000 characters.
• size controls the longest subject dimension in meters.
• Exact likeness, physical accuracy, and photorealism are not guaranteed.
• Lighting and material appearance may vary by 3D viewer.

Related Models

• GPT-6 Astra Text to 3D — Generate a textured 3D object or scene directly from a text prompt.

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'
{
  "images": [
    "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
  ],
  "size": 1
}
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/openai/gpt-6-astra/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