# TokenRig

> Run TokenRig by infsh online or via API. GPU time, billed per millisecond, no setup.

**URL:** https://inference.sh/apps/infsh/tokenrig
**Last updated:** 2026-10-28

---

Automatic rigging for 3D models with TokenRig (SkinTokens). Predicts a skeleton and skin weights in one pass and keeps the original textures. GLB, glTF, FBX, OBJ in; GLB or FBX out.

- app: `infsh/tokenrig`
- maker: [infsh](https://inference.sh/apps/infsh.md)
- category: [3d](https://inference.sh/apps/category/3d.md)
- tasks: [3d rigging](https://inference.sh/apps/tag/3d-rigging.md)
- price: GPU time, billed per millisecond
- runs via: api, javascript and python sdk, belt cli, mcp

## tokenrig API Guide

### About
automatic rigging for 3d models with tokenrig (skintokens). predicts a skeleton and skin weights in one pass and keeps the original textures. glb, gltf, fbx, obj in; glb or fbx out.

### Get an API Key
Create one in [settings → api keys](https://app.inference.sh/settings/keys). The examples read it from `INFERENCE_API_KEY`.

```bash
export INFERENCE_API_KEY="inf_your_key"
```

### 1. Calling the API

#### Install the Client

```bash
pip install inferencesh
```

#### Run and Get Result
Submit a request and wait for the final result.

```python
import os
from inferencesh import inference

client = inference(api_key=os.environ["INFERENCE_API_KEY"])

## version @3nxf599s pinned
result = client.run({
        "app": "infsh/tokenrig@3nxf599s",
        "input": {
            "model": "https://example.com/file.png"
        }
    })

print(result["output"])
```

#### Stream Live Updates
Get real-time progress updates as the task runs.

```python
import os
from inferencesh import inference

client = inference(api_key=os.environ["INFERENCE_API_KEY"])

## version @3nxf599s pinned
## stream=True yields updates as they arrive
for update in client.run({
        "app": "infsh/tokenrig@3nxf599s",
        "input": {
            "model": "https://example.com/file.png"
        }
    }, stream=True):
    if update.get("progress"):
        print(f"progress: {update['progress']}%")
    if update.get("output"):
        print(f"output: {update['output']}")
```

#### Run in the Background
Start the task and get its id back right away, then check on it with `GET /tasks/{id}`. It is done when `status` is 10 (completed), 11 (failed) or 12 (cancelled).

```python
import os
from inferencesh import inference, TaskStatus

client = inference(api_key=os.environ["INFERENCE_API_KEY"])

## returns as soon as the task is created
task = client.run({
        "app": "infsh/tokenrig@3nxf599s",
        "input": {
            "model": "https://example.com/file.png"
        }
    }, wait=False)

## later: read its current state
state = client.tasks.get(task["id"]).data
if state["status"] == TaskStatus.COMPLETED:
    print(state["output"])
```

### 2. Files

#### Automatic Upload

```python
## local file paths are automatically uploaded
result = client.run({
    "app": "infsh/tokenrig@3nxf599s",
    "input": {
        "image": "/path/to/local/image.png",  # detected & uploaded
        "audio": "https://example.com/audio.mp3",  # url passed through
    }
})
```

#### Manual Upload

```python
## upload and get a hosted URL
file = client.files.upload("/path/to/file.png")
print(file.uri)  # https://cloud.inference.sh/...
```

### 3. Schema

#### Input
- **model** (string) *required*: The 3D model to rig: GLB, glTF, FBX, OBJ, DAE or .blend.
- **output_format** (string): File format of the rigged model.
- **keep_original** (boolean): Put the rig onto the original model, keeping its textures and materials. Off exports the mesh the model rigged.
- **smooth_skin** (boolean): Clean up the skin weights with a voxel pass.
- **num_beams** (integer): Beam search width. Higher is slower and often better.
- **top_k** (integer): Sampling top-k.
- **top_p** (number): Sampling top-p.
- **temperature** (number): Sampling temperature.
- **repetition_penalty** (number): Repetition penalty.
- **seed** (integer): Seed. Empty: random.

#### Output
- **model** (string): The rigged model, with its skeleton and skin weights.

## other 3d apps

- [Tripo Auto Rig](https://inference.sh/apps/tripo/tripo-rig.md): $0.01/credit (rig check free, auto rig 25 credits, retarget 10 credits/animation)
- [Hunyuan 3D Auto Rigging](https://inference.sh/apps/tencent/hunyuan-3d-auto-rigging.md): $0.02/credit, 10 credits/call ($0.20/call)
- [Hunyuan 3D 3.1](https://inference.sh/apps/tencent/hunyuan-3d-3-1.md): $0.02/credit; Geometry 15cr, Normal 25cr, +10cr views/PBR/face-count
- [Rodin Gen-2.5](https://inference.sh/apps/hyper3d/rodin-gen-2-5.md): $0.58/credit; Extreme-High tier +0.5 credits, texture extreme-high +2 credits
- [Meshy 7.1](https://inference.sh/apps/meshy/meshy-7-1.md): $0.0127/Meshy credit (20 base + 10-15 texture + 5 hi-res geo)

## faq

### how much does TokenRig cost?

GPU time, billed per millisecond. you pay per run from your inference shell credits. no subscription is required.

### how do I run TokenRig via api?

call infsh/tokenrig with the javascript or python sdk, the rest api, or `belt app run infsh/tokenrig` from a terminal. the api reference on this page lists every input.

### who makes TokenRig?

TokenRig is made by infsh and runs on inference shell. every infsh app is listed at https://inference.sh/apps/infsh.

### can my agent use TokenRig?

yes. agents call it as a tool through the inference shell mcp server (https://api.inference.sh/mcp), the belt cli or the sdk, and pay per run like you do.

### what are the alternatives to TokenRig?

other 3d apps on inference shell include Tripo Auto Rig, Hunyuan 3D Auto Rigging, Hunyuan 3D 3.1, Rodin Gen-2.5. they run through the same api, so switching is a one-line change.