# Idol Max

> Turn one image into a game-ready 3D mesh with clean geometry and a sharp repainted texture, with an optional rig and animations.


## Overview

- **Endpoint**: `https://fal.run/speridlabs/idol/max`
- **Model ID**: `speridlabs/idol/max`
- **Category**: image-to-3d
- **Kind**: inference
**Description**: Upload one image of a single object, character or creature: Idol Max predicts a Gaussian splat, fuses its surface from stereo depth for smooth, clean geometry, builds a 30k-triangle game-ready mesh with UVs and a normal map, and repaints its base color from the image. The background is removed for you, so photographs with a scene behind the subject work. About 50 seconds per request, longer with a rig. Options: rig (adds a skeleton and skin weights, for characters and creatures), animations (walk, run, jump, idle: looping glTF clips written into the rigged mesh, requires rig), image_type (synthetic for renders and generated images, real for photographs), mesh_resolution (default 1024), num_gaussians (default 262144), num_inference_steps (default 50), guidance_scale (default 3), enable_pbr (adds roughness and metallic maps), seed. Output: game_glb (30k triangles, UVs, 2048 px base color and normal map, plus the skeleton and animations when asked), model_glb (the full-resolution vertex-colored mesh), rig (bones and animations written), preprocessed_image, mesh_stats, timings, seed.

**Tags**: image-to-3d, 3d, game-assets, rigging, animation



## Pricing

Your request will cost **$0.20** per generation, or **$0.25** when a rig is added.

For more details, see [fal.ai pricing](https://fal.ai/pricing).

## API Information

This model can be used via our HTTP API or more conveniently via our client libraries.
See the input and output schema below, as well as the usage examples.


### Input Schema

The API accepts the following input parameters:


- **`image_url`** (`string`, _required_):
  URL of the image of a single object to convert into a game-ready 3D mesh.
  - Examples: "https://v3b.fal.media/files/b/0ab21819/3xTHAXZi41yCcq25k0uZY_YGG7j3lg.png"

- **`enable_pbr`** (`boolean`, _optional_):
  Add roughness and metallic maps to the game-ready mesh with a TRELLIS.2 retexture of the preprocessed image; the base color is not changed. Adds typically 20 seconds to 2 minutes.
  - Default: `false`

- **`rig`** (`boolean`, _optional_):
  Also return `rigged_character_glb`: the game-ready mesh with a skeleton and skin weights from SkinTokens, for characters and creatures (bipeds, quadrupeds, winged or tailed). `game_glb` stays without a skeleton. Rigid props such as chests get an unusable skeleton. Adds about 15 seconds (about 60 more on a runner's first such request, which loads the model).
  - Default: `false`

- **`animations`** (`list<Enum>`, _optional_):
  Looping clips to write into `rigged_character_glb` as glTF animations; each is also returned alone in a file of the result's `animations`. Requires `rig`. Each is a hand-made animation of a human or of a four-legged animal, played in place on the limbs found from the skeleton and the skin; a clip the skeleton cannot play (walk, run and jump need two legs, or three or four on an animal) is left out, see `rig.animations` in the result.
  - Default: `[]`
  - Array of Enum

- **`image_type`** (`ImageTypeEnum`, _optional_):
  Kind of input image. `synthetic` for renders and generated images (tight crop), `real` for photographs (looser crop). Default value: `"synthetic"`
  - Default: `"synthetic"`
  - Options: `"synthetic"`, `"real"`

- **`num_gaussians`** (`integer`, _optional_):
  Target number of Gaussians in the output splat. Rounded down to a multiple of 32. Default value: `262144`
  - Default: `262144`
  - Range: `32768` to `1048576`

- **`num_inference_steps`** (`integer`, _optional_):
  Number of sampling steps. Fewer are faster but lose some detail. Default value: `50`
  - Default: `50`
  - Range: `1` to `50`

- **`guidance_scale`** (`float`, _optional_):
  Classifier-free guidance: how strictly the object follows the image. Default value: `3`
  - Default: `3`
  - Range: `0` to `10`

- **`seed`** (`integer`, _optional_):
  Random seed for reproducibility. If None, a random seed is generated.

- **`enable_safety_checker`** (`boolean`, _optional_):
  Enable safety checking of the input image. Default value: `true`
  - Default: `true`

- **`mesh_resolution`** (`MeshResolutionEnum`, _optional_):
  Surface detail. The surface is fused from rendered splat depth on a grid of 0.75 x this many voxels along the object's longest side (1024 -> 768), capped near 80M voxels for bulky objects. Lower values are faster and coarser; above 1024 mostly adds time. Default value: `"1024"`
  - Default: `1024`
  - Options: `128`, `256`, `512`, `768`, `1024`, `1536`



**Required Parameters Example**:

```json
{
  "image_url": "https://v3b.fal.media/files/b/0ab21819/3xTHAXZi41yCcq25k0uZY_YGG7j3lg.png"
}
```

**Full Example**:

```json
{
  "image_url": "https://v3b.fal.media/files/b/0ab21819/3xTHAXZi41yCcq25k0uZY_YGG7j3lg.png",
  "animations": [],
  "image_type": "synthetic",
  "num_gaussians": 262144,
  "num_inference_steps": 50,
  "guidance_scale": 3,
  "enable_safety_checker": true,
  "mesh_resolution": 1024
}
```


### Output Schema

The API returns the following output format:

- **`rigged_character_glb`** (`File`, _optional_):
  The game-ready mesh with its skeleton, skin weights and all the requested animations in one file (.glb). Only with `rig`, when the object could be rigged.

- **`animations`** (`list<RigAnimation>`, _optional_):
  One file per requested animation that was written, in the order of the request: the rigged mesh with that clip alone. Empty without `rig`, when the object could not be rigged or when no animation was requested.
  - Default: `[]`
  - Array of RigAnimation

- **`game_glb`** (`File`, _required_):
  Game-ready mesh (.glb), without a skeleton: 30k triangles, UVs, base color and normal map at 2048 px. On /max the base color is repainted from the image; on /max/turbo it is projected from the splat. With `enable_pbr`, also a metallic/roughness map.

- **`model_glb`** (`File`, _required_):
  Full-resolution vertex-colored triangle mesh (.glb), upright in glTF +Y-up coordinates.

- **`rig`** (`RigStats`, _optional_):
  The skeleton of `rigged_character_glb`, added by `rig`, and the clips written into that file; absent without `rig`.

- **`preprocessed_image`** (`Image`, _required_):
  The background-removed, cropped image the model saw.

- **`num_gaussians`** (`integer`, _required_):
  Number of Gaussians in the intermediate splat.

- **`mesh_stats`** (`MeshStats`, _required_):
  Triangle mesh geometry and closure information.

- **`seed`** (`integer`, _required_):
  The seed used for generation.

- **`timings`** (`Timings`, _required_):
  Per-stage timings in seconds.



**Example Response**:

```json
{
  "animations": [],
  "game_glb": {
    "url": "",
    "content_type": "image/png",
    "file_name": "z9RV14K95DvU.png",
    "file_size": 4404019
  },
  "model_glb": {
    "url": "",
    "content_type": "image/png",
    "file_name": "z9RV14K95DvU.png",
    "file_size": 4404019
  },
  "preprocessed_image": {
    "url": "",
    "content_type": "image/png",
    "file_name": "z9RV14K95DvU.png",
    "file_size": 4404019,
    "width": 1024,
    "height": 1024
  },
  "mesh_stats": {}
}
```


## Usage Examples

### cURL

```bash
curl --request POST \
  --url https://fal.run/speridlabs/idol/max \
  --header "Authorization: Key $FAL_KEY" \
  --header "Content-Type: application/json" \
  --data '{
     "image_url": "https://v3b.fal.media/files/b/0ab21819/3xTHAXZi41yCcq25k0uZY_YGG7j3lg.png"
   }'
```

### Python

Ensure you have the Python client installed:

```bash
pip install fal-client
```

Then use the API client to make requests:

```python
import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
           print(log["message"])

result = fal_client.subscribe(
    "speridlabs/idol/max",
    arguments={
        "image_url": "https://v3b.fal.media/files/b/0ab21819/3xTHAXZi41yCcq25k0uZY_YGG7j3lg.png"
    },
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)
```

### JavaScript

Ensure you have the JavaScript client installed:

```bash
npm install --save @fal-ai/client
```

Then use the API client to make requests:

```javascript
import { fal } from "@fal-ai/client";

const result = await fal.subscribe("speridlabs/idol/max", {
  input: {
    image_url: "https://v3b.fal.media/files/b/0ab21819/3xTHAXZi41yCcq25k0uZY_YGG7j3lg.png"
  },
  logs: true,
  onQueueUpdate: (update) => {
    if (update.status === "IN_PROGRESS") {
      update.logs.map((log) => log.message).forEach(console.log);
    }
  },
});
console.log(result.data);
console.log(result.requestId);
```


## Deploy your own model

To run your own weights, a fine-tune or a custom pipeline on the GPUs that serve this API, deploy it on fal Serverless. GPUs from H100 (80GB) to B300 (288GB), billed per second; an H100 is $4.50/h list, as low as $2.49/h. Prices, GPU specs and deploy steps: [fal Serverless](https://fal.ai/serverless.md).

## Additional Resources

### Documentation

- [Model Playground](https://fal.ai/models/speridlabs/idol/max)
- [API Documentation](https://fal.ai/models/speridlabs/idol/max/api)
- [OpenAPI Schema](https://fal.ai/api/openapi/queue/openapi.json?endpoint_id=speridlabs/idol/max)

### fal.ai Platform

- [Platform Documentation](https://fal.ai/docs/documentation)
- [Python Client](https://fal.ai/docs/api-reference/client-libraries/python)
- [JavaScript Client](https://fal.ai/docs/api-reference/client-libraries/javascript)

### Other agent-readable surfaces

This file covers one model. To find anything else:

- [Platform overview](https://fal.ai/llms.txt): Entry points and representative endpoint IDs
- [Documentation index](https://fal.ai/docs/llms.txt): Every documentation page
- [Full documentation text](https://fal.ai/docs/llms-full.txt): The whole documentation inlined
- [Pricing](https://fal.ai/pricing.md): Model API and Serverless GPU prices
- [fal Serverless](https://fal.ai/serverless.md): Deploy your own models: GPU prices, specs and steps
- Any other model: `https://fal.ai/models/<endpoint-id>/llms.txt`
