Meshy 6 Multi Image to 3D API

meshy/v6/multi-image-to-3d

Meshy 6 Multi Image to 3D builds a 3D model from one to four views of the same object with topology and texture controls. Provide complementary angles, then adjust remeshing and material guidance for the generated mesh.

Input

Image UrlsRequired

Drag and drop files, paste from clipboard, or provide a URL. Provide 1–4 publicly accessible HTTP(S) image URLs showing the same object from different angles.

3/4 files added

TopologyOptional

Target PolycountOptional

Should RemeshOptional

Symmetry ModeOptional

Should TextureOptional

Advanced settings

Enable PbrOptional

Pose ModeOptional

Texture PromptOptional

Texture Image UrlOptional

Enable RiggingOptional

Rigging Height MetersOptional

Enable AnimationOptional

Animation Action IdOptional

Enable Safety CheckerOptional

Output

Meshy 6 Multi Image to 3D model ready

Loading 3D preview…

Continue using

Meshy 6 Text to 3D Meshy 6 Image to 3D

Examples

Exposed-Joinery Elm Stool

Walnut Hand-Crank Organ

Meshy 6 Multi Image to 3D API

Quick Start

Submit Meshy 6 Multi Image to 3D inputs, retain task_id, and query the task for the returned files.

Authenticate for Meshy 6 Multi Image to 3D

Send your Vidgo API key in the Bearer authorization header.

Endpoint: POSThttps://api.vidgo.ai/api/generate/submit
Authentication: Authorization: Bearer YOUR_API_KEY

Submit a Meshy 6 Multi Image to 3D request

Place model at the request root and the image_urls value inside input.

Request example

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "meshy/v6/multi-image-to-3d",
  "input": {
    "image_urls": [
      "https://example.com/object-front.png",
      "https://example.com/object-side.png"
    ],
    "topology": "triangle",
    "target_polycount": 30000,
    "should_remesh": true,
    "symmetry_mode": "auto"
  }
}
JSON
)

RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

CODE=$(printf '%s' "$RESPONSE" | jq -r '.code // empty')
if [ "$CODE" != "0" ] && [ "$CODE" != "200" ]; then
  printf 'API error: %s
' "$RESPONSE" >&2
  exit 1
fi
printf '%s
' "$RESPONSE"

Check the Meshy 6 Multi Image to 3D task

Query the returned task_id until status is finished or failed.

Status endpoint

GET https://api.vidgo.ai/api/generate/status/{task_id}

When status is finished, inspect files and download the returned assets.

not_startedrunningfinishedfailed

Submission response

{
  "code": 200,
  "data": {
    "task_id": "your_task_id",
    "status": "not_started",
    "created_time": "2026-09-27T10:00:00"
  }
}

Completed response

{
  "code": 200,
  "data": {
    "task_id": "your_task_id",
    "status": "finished",
    "files": [
      {
        "file_url": "https://example.com/model.glb",
        "file_type": "3d",
        "label": "model_glb",
        "format": "glb",
        "file_name": "model.glb"
      },
      {
        "file_url": "https://example.com/preview.png",
        "file_type": "image",
        "label": "thumbnail",
        "format": "png"
      }
    ],
    "progress": 100,
    "error_message": null
  }
}

Read Meshy 6 Multi Image to 3D files

Read each returned files item for file_url, file_type, label, and format.

Read Meshy 6 Multi Image to 3D files

set -euo pipefail
: "${VIDGO_API_KEY:?Set VIDGO_API_KEY in your environment}"

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "meshy/v6/multi-image-to-3d",
  "input": {
    "image_urls": [
      "https://example.com/object-front.png",
      "https://example.com/object-side.png"
    ],
    "topology": "triangle",
    "target_polycount": 30000,
    "should_remesh": true,
    "symmetry_mode": "auto"
  }
}
JSON
)

SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

TASK_ID=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.data.task_id // .task_id // empty')
BUSINESS_CODE=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.code // empty')
if [ "$BUSINESS_CODE" != "0" ] && [ "$BUSINESS_CODE" != "200" ]; then
  printf 'Submit failed:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi
if [ -z "$TASK_ID" ]; then
  printf 'Submit response did not include task_id:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi

START_TIME=$(date +%s)
POLL_DELAY=2
while true; do
  if [ $(( $(date +%s) - START_TIME )) -ge 600 ]; then
    printf 'Timed out after 600 seconds
' >&2
    exit 1
  fi
  STATUS_RESPONSE=$(curl --silent --show-error --fail-with-body \
    --url "https://api.vidgo.ai/api/generate/status/$TASK_ID" \
    --header "Authorization: Bearer $VIDGO_API_KEY")
  STATUS=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.data.status // .status // empty')
  BUSINESS_CODE=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.code // empty')
  if [ "$BUSINESS_CODE" != "0" ] && [ "$BUSINESS_CODE" != "200" ]; then
    printf 'Status request failed:
%s
' "$STATUS_RESPONSE" >&2
    exit 1
  fi

  case "$STATUS" in
    finished)
      printf '%s' "$STATUS_RESPONSE" | jq -r '(.data.files // .files // [])[]?.file_url'
      break
      ;;
    failed)
      printf '%s' "$STATUS_RESPONSE" | jq -r '.data.error_message // .error_message // "Generation failed"' >&2
      exit 1
      ;;
    not_started|running)
      sleep "$POLL_DELAY"
      if [ "$POLL_DELAY" -lt 10 ]; then POLL_DELAY=$((POLL_DELAY + 1)); fi
      ;;
    *)
      printf 'Unexpected task status: %s
' "$STATUS" >&2
      exit 1
      ;;
  esac
done

Input Parameters

Place model and optional callback_url at the request root; put the fields listed below inside input.

Field	Type	Required	Default	Description
model	string	Yes	—	meshy/v6/multi-image-to-3d
input	object	Yes	—	Generation settings for this endpoint.
input.image_urls	array	Yes	—	Provide 1–4 publicly accessible HTTP(S) image URLs showing the same object from different angles.
input.topology	string	No	triangle	Choose triangle or quad mesh topology. Default: triangle.
input.target_polycount	integer	No	30000	Set the target polygon count as an integer. Default: 30000.
input.should_remesh	boolean	No	true	Enable remeshing to apply the chosen topology and polygon target. Default: true.
input.symmetry_mode	string	No	auto	Choose off, auto, or on for symmetry. Default: auto.
input.should_texture	boolean	No	true	Generate textures from the reference images. Default: true.
input.enable_pbr	boolean	No	—	Generate PBR material maps for metallic, roughness, and normal details.
input.pose_mode	string	No	""	Choose a-pose, t-pose, or an empty string for the model pose. Default: empty string.
input.texture_prompt	string	No	—	Describe colors, patterns, and materials to guide the texture.
input.texture_image_url	string	No	—	Provide a publicly accessible HTTP(S) image URL as a texture reference.
input.enable_rigging	boolean	No	—	Add a skeleton to a humanoid character for posing and animation.
input.rigging_height_meters	number	No	1.7	Set the character height in meters when rigging. Default: 1.7.
input.enable_animation	boolean	No	—	Apply an animation preset with enable_rigging set to true.
input.animation_action_id	integer	No	1001	Select an integer animation preset ID. Default: 1001.
input.enable_safety_checker	boolean	No	true	Enable input safety checking. Default: true.
callback_url	string	No	—	Provide a public HTTP(S) URL for completed or failed task notifications.

Response Fields

The status response reports task state, progress, and returned files.

Field	Type	Description
code	integer	API response code.
data.task_id	string	Task ID for status queries.
data.status	string	Task state: not_started, running, finished, or failed.
data.progress	integer	Reported generation progress from 0 to 100.
data.files	array	Generated model, preview image, and companion files.
data.files[].file_url	string	URL for a returned file.
data.files[].file_type	string	File category: 3d, image, or other.
data.files[].label	string	File role, such as model_glb or thumbnail.
data.files[].format	string	File format, such as glb or png.
data.files[].file_name	string	File name when provided.
data.files[].file_size	integer	File size in bytes when provided.
data.error_message	string \| null	Error details for a failed task.

Task Lifecycle

Use task_id to check the same generation until it finishes or fails.

not_started
The task has been accepted and awaits generation.
running
The 3D model is being generated; continue querying this task.
finished
Generation is complete; read files for the assets.
failed
The task failed; read error_message before submitting another request.

Polling and Errors

Keep the task IDIf a status query is interrupted, use the same task_id to retrieve the latest state.
Download returned assetsUse file_url and file_name from each returned file to download the model and companion assets.
Review a failed generationWhen status is failed, read error_message, then review the input before submitting a new task.

Endpoint Specifications

Item	Value	Description
Input	one to four images of the same object	Provide 1–4 publicly accessible HTTP(S) images showing the same object from different angles.
Topology	triangle, quad	Choose triangle or quad mesh topology. Default: triangle.
Output files	GLB, FBX, OBJ, USDZ, BLEND, STL	Read the files returned by the task for model formats, textures, and animation assets.

View the Meshy 6 Multi Image to 3D API docs

Meshy 6 Multi Image to 3D

Meshy 6 Multi Image to 3D is a Meshy model that builds a 3D asset from one to four images of the same object. Different angles supply complementary shape references, while topology, polygon targets, symmetry, and remeshing control the resulting mesh. Texture generation, PBR maps, and surface references support material work; humanoid assets can also use pose, rigging, and animation settings.

Why Choose Meshy 6 Multi Image to 3D?

Combine up to four object viewsSubmit one to four images of the same object from different angles as shape references.
Shape the multiview meshChoose triangle or quad topology, set a polygon target, and adjust remeshing and symmetry.
Develop surfaces and humanoid assetsGenerate textures from the references, add PBR maps, or rig a humanoid model for an animation preset.

Parameters

Parameter	Requirement	What it controls
image_urls	Required	Provide 1–4 publicly accessible HTTP(S) image URLs showing the same object from different angles.
topology	Optional	Choose triangle or quad mesh topology. Default: triangle. Default`trianglequad`
target_polycount	Optional	Set the target polygon count as an integer. Default: 30000. Default`30000`
should_remesh	Optional	Enable remeshing to apply the chosen topology and polygon target. Default: true. Default`true`
symmetry_mode	Optional	Choose off, auto, or on for symmetry. Default: auto. Default`autooffon`
should_texture	Optional	Generate textures from the reference images. Default: true. Default`true`
enable_pbr	Optional	Generate PBR material maps for metallic, roughness, and normal details.
pose_mode	Optional	Choose a-pose, t-pose, or an empty string for the model pose. Default: empty string. Default`""a-poset-pose`
texture_prompt	Optional	Describe colors, patterns, and materials to guide the texture.
texture_image_url	Optional	Provide a publicly accessible HTTP(S) image URL as a texture reference.
enable_rigging	Optional	Add a skeleton to a humanoid character for posing and animation.
rigging_height_meters	Optional	Set the character height in meters when rigging. Default: 1.7. Default`1.7`
enable_animation	Optional	Apply an animation preset with enable_rigging set to true.
animation_action_id	Optional	Select an integer animation preset ID. Default: 1001. Default`1001`
enable_safety_checker	Optional	Enable input safety checking. Default: true. Default`true`

How to Use

Gather views of the same objectUpload one to four images showing the same object from different angles.
Choose mesh settingsSelect Topology and Target Polycount, then set remeshing and symmetry controls.
Choose texture controlsSet Should Texture for image-based texturing; add Texture Prompt, Texture Image Url, or Enable Pbr for materials.
Generate and inspectReview the 60-credit cost and select Run. Rotate the returned GLB model and download the available files.

Pricing

Meshy 6 Multi Image to 3D costs $0.300, or 60 credits, for each generation.

Generation	Cost	Billing unit
Meshy 6 Multi Image to 3D	$0.300 / generation · 60 credits	One submitted generation

Use Cases

Multi-angle product referencesUse several views of one product to give the generator more shape information.
Props with visible side detailsCombine front, side, and back references when an object has distinct details on multiple sides.
Humanoid character referencesUse different views of the same character with pose, skeleton, and animation controls.

Pro Tips

Show the same object in each image and vary the angle to add shape information.
Keep the object visible and framed similarly across views.
Use should_texture=true when generating textures from the references.
Set topology and target polygon count before applying remeshing.

Notes

Use the output file labels to identify the model, thumbnail, textures, and any rigging or animation files returned for the task.

Meshy 6 Multi Image to 3D Related Models

Meshy 6 Text to 3DMeshy 6 Text to 3D creates a 3D model from an object description.

Meshy 6 Image to 3DMeshy 6 Image to 3D uses one object image to generate a 3D model.

Meshy 6 Multi Image to 3D API frequently asked questions

What is the Meshy 6 Multi Image to 3D API?

Meshy 6 Multi Image to 3D is a Meshy model for creating 3D assets from one to four images of the same object. It combines different views with topology, remeshing, symmetry, and texture controls to build the model. PBR material maps and optional humanoid rigging and animation settings extend asset preparation. You can submit a request through the API or try the multiview workflow in the Playground on this page.

How many views can Meshy 6 Multi Image to 3D use?

Submit one to four publicly accessible image URLs showing the same object. Different angles can supply complementary shape references.

Which angles suit Meshy 6 Multi Image to 3D?

Use available front, side, back, or three-quarter views of the same object. Select angles that reveal the parts you want represented in the mesh.

How should Meshy 6 Multi Image to 3D frame its references?

Keep the full object visible and use similar framing across the images. Make sure handles, thin parts, and other shape details can be seen in the relevant views.

How does Meshy 6 Multi Image to 3D remesh an object?

Choose triangle or quad topology, set an integer target_polycount, and use should_remesh to apply those settings. symmetry_mode offers off, auto, and on.

How can Meshy 6 Multi Image to 3D guide textures?

Set should_texture to true to generate textures from the reference set. Use texture_prompt for written surface details, texture_image_url for an image reference, and enable_pbr for material maps.

Can Meshy 6 Multi Image to 3D rig a character?

Set enable_rigging to true to add a skeleton to a humanoid model. Set enable_animation to true with rigging enabled and choose an animation_action_id preset.

When should I choose Meshy 6 Multi Image to 3D over Image to 3D?

Use Multi Image to 3D when you have several angles of one object to supply together. Meshy 6 Image to 3D takes one reference image.