3D

TripoSG Image to 3D

Turn one photo into a 3D mesh you can download, including on a phone.

TripoSG by VAST-AI-Research; DINOv2 by Meta AI; U2-Net by Xuebin Qin and contributors; rembg by Daniel Gatis

Real-ESRGAN by Xintao Wang

4x-UltraSharp by Kim2091, CC BY-NC-SA 4.0; converted to ONNX by Mesh, unchanged learned tensors.

Model licenses

TripoSG DiT and VAE ONNX derivatives · MIT

Preserve the complete MIT terms, copyright notice and disclaimer, and identify modifications. No upstream endorsement is granted.

Distributed license

Notice

DINOv2 Large conditioning image encoder · Apache-2.0

Preserve the complete Apache-2.0 terms, copyright notices and disclaimers, identify modifications, and respect Apache patent/trademark provisions. No upstream endorsement is granted.

U2-Net matte, rembg ONNX release · Apache-2.0 AND MIT

Preserve the complete applicable MIT/Apache-2.0 terms, copyright notices and disclaimers, identify modifications, and respect Apache patent/trademark provisions. No upstream endorsement is granted.

Real-ESRGAN x4plus original release ONNX derivative · BSD-3-Clause

Retain the original copyright notice, all three BSD conditions and warranty disclaimer with distributed copies. No endorsement using author/contributor names without permission.

Distributed license

Notice

Original 4x-UltraSharp ONNX conversion by Mesh · CC-BY-NC-SA-4.0

NonCommercial only. Preserve attribution, full license and modification notice. ShareAlike applies to adapted licensed material. Private compute only, unmetered, requester and operator acceptance required; no global-network execution. Acceptance does not grant commercial permission.

Distributed license

Notice

FreeOnlineUnlimitedUltraSharp API access

API reference

Submit via REST

Use your personal API key (msk_…) to create jobs for this workload from your own tools. Jobs started this way run on the same devices you connected on Your Devices — at least one has to be online.

Hide API details
POSThttps://cp.meshlabs.cloud/v1/organizations/jobs

Authenticate with X-Organization-Api-Key: msk_your_personal_api_key. The header name matches org keys; personal keys use the same endpoint.

Input files:Call POST https://cp.meshlabs.cloud/v1/organizations/job-inputs/presign with the same API key, upload your file to the returned upload_url, then pass download_url in job_spec (for example as input_image_url).

Request body

{
  "workload_type": "triposg-1-5b-i2m",
  "job_spec": {
    "input_image_url": "https://…/your-input.png",
    "remove_background": true,
    "steps": 50,
    "guidance_scale": 7,
    "grid_resolution": 192,
    "texture": true,
    "texture_upscale": false,
    "texture_upscaler": "ultrasharp"
  }
}

job_spec fields

FieldRequiredDefaultDescription
input_image_urlYes—A photo or render of a single object. The subject is cut out automatically, cropped square and resized to 224x224, so a single clear subject gives the best geometry — busy scenes and multiple objects confuse the cutout. A PNG that already has transparency skips the cutout step and is the most reliable input of all.
remove_backgroundNotrueRun the segmentation model when the image has no transparency of its own. Turn this off for an image that is already a clean cutout on white, or when the subject fills the frame.
stepsNo50Rectified-flow steps. Below about 16 the shape is recognisable but loses fine relief; above the default it stops changing.
guidance_scaleNo7How closely the mesh follows the reference image.
seedNo—Random seed for reproducible generation.
grid_resolutionNo192Resolution of the field the surface is extracted from. Cost grows with the cube: 256 is roughly 2.4x the work of the default for a difference that mostly does not survive smoothing.
textureNotrueProject the photo onto the mesh as a base-colour texture. The side the photo showed comes out sharp; the far side is a smear of the nearest silhouette colour, because one image cannot see it. Turn this off for a plain grey mesh — smaller file, and the right choice if you are going to texture it yourself.
texture_upscaleNofalseRun a 4x super-resolution pass over the subject before projecting it. Only worth it when the subject is small in the frame — the texture is built at the photo's own resolution, so a large clear photo already has more detail than it holds and this would only upscale our own downscale. It is skipped automatically when the photo is big enough, so leaving it on costs nothing but a download.
texture_upscalerNoultrasharpUltraSharp: preferred detail, private NonCommercial use with terms acceptance. x4plus: BSD-3-Clause alternative. Used only when sharpening a small photo is enabled.

cURL

curl -X POST https://cp.meshlabs.cloud/v1/organizations/jobs \
  -H "X-Organization-Api-Key: msk_your_personal_api_key" \
  -H "Content-Type: application/json" \
  -d '{
  "workload_type": "triposg-1-5b-i2m",
  "job_spec": {
    "input_image_url": "https://…/your-input.png",
    "remove_background": true,
    "steps": 50,
    "guidance_scale": 7,
    "grid_resolution": 192,
    "texture": true,
    "texture_upscale": false,
    "texture_upscaler": "ultrasharp"
  }
}'
GEThttps://cp.meshlabs.cloud/v1/organizations/jobs/{job_id}

Poll with the same API key until status is terminal (COMPLETED, FAILED, CANCELED, or MODERATION_BLOCKED). When content moderation is enabled, jobs may start in MODERATION_PENDING before entering the compute queue. Completed jobs include presigned download_url values on each output.

Create response

{
  "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "workload_type": "triposg-1-5b-i2m",
  "status": "PENDING",
  "job_spec_json": {
    "input_image_url": "https://…/your-input.png",
    "remove_background": true,
    "steps": 50,
    "guidance_scale": 7,
    "grid_resolution": 192,
    "texture": true,
    "texture_upscale": false,
    "texture_upscaler": "ultrasharp"
  },
  "work_units": [],
  "outputs": [],
  "created_at": "2026-06-08T12:00:00Z",
  "updated_at": "2026-06-08T12:00:00Z"
}

Completed response

{
  "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "workload_type": "triposg-1-5b-i2m",
  "status": "COMPLETED",
  "job_spec_json": {
    "input_image_url": "https://…/your-input.png",
    "remove_background": true,
    "steps": 50,
    "guidance_scale": 7,
    "grid_resolution": 192,
    "texture": true,
    "texture_upscale": false,
    "texture_upscaler": "ultrasharp"
  },
  "work_units": [
    {
      "id": "…",
      "index": 0,
      "status": "COMPLETED"
    }
  ],
  "outputs": [
    {
      "id": "…",
      "output_type": "<output type>",
      "filename": "<output filename>",
      "size_bytes": 524288,
      "download_url": "<presigned output download URL>"
    }
  ],
  "created_at": "2026-06-08T12:00:00Z",
  "updated_at": "<completion timestamp>"
}

Upload one photo and download a 3D mesh (GLB). The subject is cut out of the background. Colour comes from the photo: the side you photographed is sharp, the far side is softer, and there are no metallic or roughness maps. Free image to 3D on devices you connect, including a phone. The first run downloads the model; later photos start faster. Advanced options turn the cutout and the texture off, and turn a larger texture on when the subject is too small in the frame.