Meshy V7 Text To 3d API Documentation

Meshy V7 Text To 3d API Documentation

Playground

Meshy V7 Text-to-3D turns text prompts into textured, PBR-ready 3D models with configurable geometry, topology, rigging, and animation for game assets, product visualization, 3D design, and production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Meshy V7 Text to 3D generates a 3D model directly from a text prompt. Describe the object you want to create, choose preview or full mode, configure geometry settings, and optionally enable PBR materials, Ultra Mode, and humanoid rigging.

Why Choose This?

• Text-to-3D generation
  Create a 3D model from a written object description.

• Preview and full modes
  Use preview for untextured geometry or full for a textured model.

• Flexible geometry controls
  Configure topology, target polygon count, remeshing, symmetry, and model type.

• Multiple model types
  Choose standard, lowpoly, or smart-topology depending on your downstream workflow.

• Texture guidance in full mode
  Use texture_prompt or texture_image to guide texture appearance when generating a full textured model.

• Optional PBR and rigging
  Generate PBR material maps in full mode and optionally rig suitable humanoid characters.

Parameters

How to Use

1. Write your prompt — Describe the object, shape, material, style, and important visual details.
2. Choose mode — Use preview for quick untextured geometry or full for textured output.
3. Select model type — Choose standard, lowpoly, or smart-topology.
4. Configure geometry optional — Adjust topology, polygon count, remeshing, and symmetry settings.
5. Add texture guidance optional — Use texture_prompt or texture_image when generating in full mode.
6. Enable PBR optional — Use enable_pbr when metallic, roughness, and normal maps are needed.
7. Enable Ultra Mode optional — Use ultra_mode for higher-fidelity standard geometry.
8. Enable rigging optional — Use enable_rigging for suitable humanoid characters.
9. Set seed optional — Use a fixed seed for reproducible results, or -1 for random generation.
10. Submit — Generate the 3D model and retrieve the output through the standard WaveSpeed prediction response.

Pricing

Pricing depends on selected mode, ultra_mode, and whether enable_rigging is enabled.

Example Costs

enable_pbr, model_type, topology, target_polycount, should_remesh, symmetry_mode, pose_mode, texture_prompt, texture_image, and seed do not add separate charges in the current pricing formula.

Best Use Cases

• Text-to-3D asset creation — Generate 3D models directly from object descriptions.
• Product and object prototypes — Turn product ideas or concept descriptions into 3D assets.
• Game and real-time workflows — Use lowpoly or smart-topology for optimized geometry.
• Humanoid character workflows — Generate and optionally rig suitable humanoid characters.
• Material and texture exploration — Use full mode with texture prompts or texture images for styled 3D assets.
• Fast geometry previews — Use preview mode to test object structure before generating textured output.

Pro Tips

• Describe the object clearly, including shape, material, style, scale, and key details.
• Use preview mode for fast geometry checks before running full textured generation.
• Use full mode when you need textured output.
• Use texture_prompt to describe material details such as metal, leather, ceramic, fabric, plastic, or wood.
• Use texture_image when you need texture appearance to follow a visual reference.
• Use lowpoly for lightweight assets and smart-topology when cleaner optimized topology is needed.
• Keep target_polycount at or below 15000 when using smart-topology.
• Use ultra_mode only with standard geometry.
• Enable rigging only for suitable humanoid characters.
• Set a fixed seed when comparing prompt or parameter changes.

Related Models

• Meshy V7 Image to 3D — Generate a 3D model from a single reference image.
• Meshy V7 Multi-Image to 3D — Generate a 3D model from multiple reference images of the same object.

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",
  "mode": "full",
  "model_type": "standard",
  "topology": "triangle",
  "target_polycount": 30000,
  "should_remesh": true,
  "symmetry_mode": "auto",
  "enable_pbr": false,
  "pose_mode": "a-pose",
  "ultra_mode": false,
  "enable_rigging": false,
  "rigging_height_meters": 1.7
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/meshy/v7/text-to-3d" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

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

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted) 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