# Trellis 2 API

> Ginagawang textured na GLB 3D mesh ang isang larawan, na may naaayos na mga setting para sa structure, shape, texture at UV

Bumubuo ng 3D asset mula sa isang larawan (`image_url`) at nagbabalik ng GLB file. Ang `resolution` ay 512, 1024 o 1536, at ang `texture_size` ay 1024, 2048 o 4096. May sariling mga kontrol sa guidance at sampling-step ang bawat isa sa tatlong stage (structure, shape, texture). Itinatakda ng `decimation_target` ang bilang ng vertex ng mesh (5,000 hanggang 2,000,000), at available ang mga opsyon sa remeshing at UV-unwrapping.

- Model id: `trellis-2`
- Made by: Microsoft
- Category: Image to 3D
- Output: 3d
- Price: from $0.25 / request (cheapest run $0.25 (250 API credits))
- Run: `POST https://api.doitong.com/v1/run/trellis-2`
- Quote: `POST https://api.doitong.com/v1/quote/trellis-2`
- Docs page: https://doitong.com/fil/developers/models/trellis-2

## Pricing

Price: **from $0.25 / request** (the lowest tier; the quote prices your exact input). Billed once per request.

Price rules (the first rule that matches the input wins; a parameter you leave out counts as its default):

- When `resolution` = 1024: $0.30 (300 API credits) / request
- When `resolution` = 1536: $0.35 (350 API credits) / request

The full example input below costs $0.30 = 300 API credits (1 request × 300 API credits).

Prices are the upstream price in USD, paid from the API balance in API credits (1000 API credits = 1 USD). `POST https://api.doitong.com/v1/quote/trellis-2` with the same JSON body returns the exact price for that input; the quote is exactly what a run charges. Invalid input is never charged and failed or canceled requests are refunded automatically.

## Input parameters

Send either the canonical name or the native field name (not both). Canonical names are the same across models.

| Parameter | Canonical name | Type | Required | Default | Allowed values / range | Description |
|---|---|---|---|---|---|---|
| `seed` |  | integer \| null | no |  |  | Seed na ginagamit para sa reproducible na random pagbuo. |
| `resolution` |  | enum (integer) | no | `1024` | `512`, `1024`, `1536` | Resolution ng output; nagdaragdag ng detalye ang pagtaas nito ngunit mas tumatagal. |
| `ss_guidance_strength` |  | number | no | `7.5` | 0 – 10 | Kontrol sa fidelity ng paunang 3D structure: mas pinapaboran ng pagtaas nito ang mas malapit na pagtutugma sa input larawan kapalit ng posibilidad ng dagdag na noise. |
| `ss_guidance_rescale` |  | number | no | `0.7` | 0 – 1 | Kontrol sa guidance artifact ng Stage 1, kung saan mas malakas ang epekto ng guidance sa mababang setting at mas stable ang output sa mataas na setting. |
| `ss_guidance_interval_start` |  | number | no | `0.6` | 0 – 1 | Simulang boundary ng classifier-free guidance window sa stage 1 (sparse structure), na ipinapahayag bilang fraction ng denoising. Mas maagang nagsisimula ang guidance sa mas mababang setting; interval_start hanggang interval_end lamang ang gumagamit nito, at unguided ang pagproseso sa labas ng interval. |
| `ss_guidance_interval_end` |  | number | no | `1` | 0 – 1 | Cutoff ng guidance para sa stage 1 (sparse structure): pumili ng value na hindi bababa sa ss_guidance_interval_start. Tinutukoy ng value ang fraction ng denoising kung kailan hihinto ang classifier-free guidance. |
| `ss_sampling_steps` |  | integer | no | `12` | 1 – 50 | Bilang ng denoising para sa paunang structure: maaaring tumaas ang kalidad sa dagdag na step ngunit mas mabagal ang processing. |
| `ss_rescale_t` |  | number | no | `5` | 1 – 6 | Itakda ang sharpness ng transition sa noise schedule ng structure pagbuo: mas nagiging matalas ang transition kapag tinaasan ang value na ito. |
| `shape_slat_guidance_strength` |  | number | no | `7.5` | 0 – 10 | Kontrol sa fidelity ng detalyadong geometry: mas maraming detalye ang nagmumula sa mas mahigpit na pagsunod sa input larawan, ngunit may panganib ng dagdag na noise. |
| `shape_slat_guidance_rescale` |  | number | no | `0.5` | 0 – 1 | Taasan ang setting na ito kapag maingay ang geometry: binabawasan nito ang mga artifact na dulot ng malakas na guidance sa shape stage. |
| `shape_slat_guidance_interval_start` |  | number | no | `0.6` | 0 – 1 | Activation point ng classifier-free guidance para sa shape refinement sa stage 2. Bilang fraction ng progreso ng denoising, itinatakda nito ang interval_start; gumagana lamang ang guidance sa window sa pagitan ng interval_start at interval_end. |
| `shape_slat_guidance_interval_end` |  | number | no | `1` | 0 – 1 | Cutoff ng guidance para sa stage 2 (shape refinement): pumili ng value na hindi bababa sa shape_slat_guidance_interval_start. Tinutukoy ng value ang fraction ng denoising kung kailan hihinto ang classifier-free guidance. |
| `shape_slat_sampling_steps` |  | integer | no | `12` | 1 – 50 | Bilang ng denoising para sa shape refinement: maaaring maging mas makinis ang geometry ngunit bumabagal kapag nagdagdag ng step. |
| `shape_slat_rescale_t` |  | number | no | `3` | 1 – 6 | Itakda ang sharpness ng geometric detail sa noise schedule ng shape refinement; mas nagiging matalas ang mga detalye kapag tinaasan ang value na ito. |
| `tex_slat_guidance_strength` |  | number | no | `1` | 0 – 10 | Kontrol sa fidelity ng kulay ng texture: mas nagiging matingkad ang mas mahigpit na pagsunod sa mga kulay ng input larawan sa mataas na setting, ngunit maaaring magdulot ng oversaturation. |
| `tex_slat_guidance_rescale` |  | number | no | `0` | 0 – 1 | Taasan ang setting na ito para tugunan ang maingay na texture o color banding: binabawasan nito ang mga artifact na dulot ng malakas na guidance sa texture stage. |
| `tex_slat_guidance_interval_start` |  | number | no | `0.6` | 0 – 1 | Classifier-free guidance activation point para sa texture pagbuo sa stage 3. Bilang fraction ng progreso ng denoising, itinatakda nito ang interval_start; gumagana lamang ang guidance sa window sa pagitan ng interval_start at interval_end. |
| `tex_slat_guidance_interval_end` |  | number | no | `0.9` | 0 – 1 | Cutoff ng guidance para sa stage 3 (texture): pumili ng value na hindi bababa sa tex_slat_guidance_interval_start. Tinutukoy ng value ang fraction ng denoising kung kailan hihinto ang classifier-free guidance. |
| `tex_slat_sampling_steps` |  | integer | no | `12` | 1 – 50 | Bilang ng denoising step na ginagamit sa pagbuo ng mga texture. Mas tumatagal kapag tinaasan ang bilang ngunit maaaring maging mas malinis ang mga texture. |
| `tex_slat_rescale_t` |  | number | no | `3` | 1 – 6 | Itakda ang sharpness ng texture detail sa noise schedule ng texture pagbuo; mas nagiging matalas ang mga detalye kapag tinaasan ang value na ito. |
| `decimation_target` |  | integer | no | `500000` | 5,000 – 2,000,000 | Target na bilang ng vertex ng mesh; inirerekomenda ang 500k para sa pangkalahatang gamit at 20k–50k para sa web/mobile. Kapag binawasan ang target, nababawasan ang detalye ng mesh kapalit ng mas maliliit na file. |
| `texture_size` |  | enum (integer) | no | `2048` | `1024`, `2048`, `4096` | Resolution ng mesh texture para sa baking. Mas malalaki ang file kapag tinaasan ang resolution kapalit ng mas pinong detalye sa surface. |
| `remesh` |  | boolean | no | `true` |  | Topology rebuilding na lumilikha ng mas malilinis na triangle ng mesh, karaniwang nagpapahusay sa pagiging angkop para sa mga susunod na gamit gaya ng animation o 3D printing, kapalit ng mas mabagal na pagproseso. |
| `remesh_band` |  | number | no | `1` | 0 – 4 | Allowance ng galaw sa remeshing: maaaring isakripisyo ang pinong detalye sa dagdag na smoothing kapag tinaasan ang value na ito. Nililimitahan ng setting ang paggalaw ng vertex kaugnay ng orihinal na surface. |
| `remesh_project` |  | number | no | `0` | 0 – 1 | Lakas ng projection para ibalik ang mga remeshed vertex sa orihinal na surface: pinananatili ng 1 ang detalye sa pamamagitan ng full projection, habang hindi gumagamit ng projection ang 0 para sa mas makinis na resulta. |
| `uv_unwrap_angle_threshold_deg` |  | number | no | `90` | 0 – 180 | Limitasyon ng anggulo ng UV chart sa degrees, na naglilimita sa pagkakaiba ng mga normal ng magkatabing face sa loob ng isang chart. Mas malalaki at mas kaunting chart ang pinapaboran ng mas mataas na limit, ngunit may panganib ng stretching; mas maraming chart at mas kaunting distortion ang pinapaboran ng mas mababang limit. |
| `uv_unwrap_refine_iterations` |  | integer | no | `0` | 0 – 10 | Bilang ng local refinement pass na ginagamit sa pag-compute ng mga UV chart. |
| `uv_unwrap_global_iterations` |  | integer | no | `1` | 1 – 10 | Mapapahusay ang balanse ng UV chart sa pamamagitan ng pagtaas sa bilang ng global clustering pass na ginagamit sa pag-compute ng mga chart. |
| `uv_unwrap_smooth_strength` |  | number | no | `1` | 0 – 10 | Ayusin kung gaano kalakas ang smoothing sa mga edge ng UV chart; ang mas hindi kapansin-pansing seam sa mas mataas na setting ay may panganib ng UV distortion. |
| `image_url` | `image` | string (image URL) | yes |  |  | URL na tumuturo sa source larawan para sa 3D conversion. |

File parameters (`image`) take a public `https://` URL, a `data:` URI up to 10 MB, or the `file_url` of an upload from `POST /files`.

## Example input

Required parameters only:

```json
{
  "image": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
}
```

A fuller example:

```json
{
  "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
}
```

## Run it

Set `DOITONG_API_KEY` to a key from https://doitong.com/developers/dashboard#keys. `?wait=20` returns the finished request in the same call when it is ready within 20 seconds; otherwise poll `GET /requests/{id}` (it also accepts `?wait=`), or pass `webhook_url`.

### curl

```bash
# Run the model; wait up to 20 s for the finished result in the same call
curl -X POST "https://api.doitong.com/v1/run/trellis-2?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
  }'

# Still "queued" or "running"? Poll with the id from the response:
curl "https://api.doitong.com/v1/requests/REQUEST_ID?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY"
```

### JavaScript

```javascript
const API = "https://api.doitong.com/v1";
const headers = {
  "Authorization": "Bearer " + process.env.DOITONG_API_KEY,
  "Content-Type": "application/json",
};

// Run the model; wait up to 20 s for the finished result in the same call
let res = await fetch(API + "/run/trellis-2?wait=20", {
  method: "POST",
  headers,
  body: JSON.stringify({
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
  }),
});
let request = await res.json();
if (!res.ok) throw new Error(request.error.code + ": " + request.error.message);

// Not finished yet: long-poll until it is
while (request.status === "queued" || request.status === "running") {
  res = await fetch(API + "/requests/" + request.id + "?wait=20", { headers });
  request = await res.json();
}
if (request.status !== "succeeded") throw new Error(JSON.stringify(request.error));
console.log(request.outputs.map((o) => o.url), request.cost.credits + " API credits ($" + request.cost.usd + ")");
```

### Python

```python
import os
import requests

API = "https://api.doitong.com/v1"
headers = {"Authorization": "Bearer " + os.environ["DOITONG_API_KEY"]}

payload = {
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg",
}

# Run the model; wait up to 20 s for the finished result in the same call
r = requests.post(API + "/run/trellis-2", params={"wait": 20}, json=payload, headers=headers)
request = r.json()
if not r.ok:
    raise RuntimeError(request["error"]["code"] + ": " + request["error"]["message"])

# Not finished yet: long-poll until it is
while request["status"] in ("queued", "running"):
    request = requests.get(API + "/requests/" + request["id"], params={"wait": 20}, headers=headers).json()
if request["status"] != "succeeded":
    raise RuntimeError(request["error"])
print([o["url"] for o in request["outputs"]], request["cost"]["credits"], "API credits", "$" + request["cost"]["usd"])
```

### MCP

The MCP server at `https://api.doitong.com/v1/mcp` exposes the tools `search_models`, `get_model`, `quote`, `run`, `get_request`, `cancel_request`, `list_requests`, `create_upload` and `get_account`. A `run` call for this model:

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "run",
    "arguments": {
      "model_id": "trellis-2",
      "input": {
        "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
      },
      "wait_seconds": 20
    }
  }
}
```

## Response

Every call returns a request object. `status` is one of `queued`, `running`, `succeeded`, `failed`, `canceled`. `outputs` lists the generated files with URLs on our storage (kept forever, or 7 days with `?storage=temporary`; `storage` says which); `output` is the model's own result object with the same URLs.

```json
{
  "id": "req_8k2m4q7x1z9c3v5b6n0p2r4t",
  "object": "request",
  "model": "trellis-2",
  "status": "succeeded",
  "created_at": "2026-10-20T12:00:00Z",
  "started_at": "2026-10-20T12:00:02Z",
  "completed_at": "2026-10-20T12:00:09Z",
  "queue_position": null,
  "cost": {
    "credits": 300,
    "usd": "0.30",
    "charged": 300,
    "refunded": 0
  },
  "input": {
    "image_url": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg"
  },
  "output": {
    "model_glb": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_name": "0-output.jpeg",
      "file_size": 1843200
    }
  },
  "outputs": [
    {
      "type": "file",
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.jpeg",
      "content_type": "image/jpeg",
      "file_size": 1843200,
      "stored": true
    }
  ],
  "error": null,
  "metrics": {
    "inference_seconds": 6.8
  },
  "urls": {
    "self": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t",
    "cancel": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t/cancel"
  }
}
```

### Output fields (`output`)

| Field | Type | Description |
|---|---|---|
| `model_glb` | object {url, content_type, file_name, file_size} | 3D modelo na nabuo sa GLB format. |

## Related models

- [Trellis 2 · Retexture](https://doitong.com/developers/models/trellis-2/retexture.md) (`trellis-2/retexture`): from $0.20 / request
- [Trellis](https://doitong.com/developers/models/trellis.md) (`trellis`): $0.02 / request
- [Trellis · Multi](https://doitong.com/developers/models/trellis/multi.md) (`trellis/multi`): $0.02 / request
- [TRELLIS.2 LoRA Inference](https://doitong.com/developers/models/trellis-2-lora.md) (`trellis-2-lora`): from $0.25 / request
- [Hi3D Image to 3D](https://doitong.com/developers/models/hitem3d/hi3d/image-to-3d.md) (`hitem3d/hi3d/image-to-3d`): $0.50 / run
- [Hi3D Image to 3D · v3.0](https://doitong.com/developers/models/hitem3d/hi3d/v3.0/image-to-3d.md) (`hitem3d/hi3d/v3.0/image-to-3d`): from $1.80 / run
- [Hi3D Multiview to 3D · Multi View to 3D](https://doitong.com/developers/models/hitem3d/hi3d/multi-view-to-3d.md) (`hitem3d/hi3d/multi-view-to-3d`): $0.50 / run
- [Hi3D Multiview to 3D · v3.0 · Multi View to 3D](https://doitong.com/developers/models/hitem3d/hi3d/v3.0/multi-view-to-3d.md) (`hitem3d/hi3d/v3.0/multi-view-to-3d`): from $1.80 / run

## Links

- HTML page: https://doitong.com/developers/models/trellis-2
- All models and guides for AI agents: https://doitong.com/developers/llms.txt
- OpenAPI 3.1 with this model's typed request body: https://doitong.com/developers/openapi.json?model=trellis-2
- Errors: https://doitong.com/developers/docs/errors
- Model details as JSON: https://api.doitong.com/v1/models/trellis-2