# Trellis2Conditioning

Trellis2Conditioning converts an input image into conditioning data for the TRELLIS.2 model. It uses a CLIP vision model to encode the image into two sets of features (at 512 and 1024 scales) and packages them as a positive conditioning pair, while also creating a matching zero-filled negative conditioning pair that serves as an empty reference.

## Inputs

| Parameter | Description | Data Type | Required | Range |
|-----------|-------------|-----------|----------|-------|
| `clip_vision_model` | The CLIP vision model used to encode the image into conditioning features. | CLIP_VISION | Yes | Any available CLIP vision model |
| `image` | Preprocessed image from ImageCropToMask (pad_factor=1.0 for TRELLIS.2). | IMAGE | Yes | Any image |

## Outputs

| Output Name | Description | Data Type |
|-------------|-------------|-----------|
| `positive` | Conditioning containing the encoded image features at 512 and 1024 scales, used as the positive conditioning for the TRELLIS.2 model. | CONDITIONING |
| `negative` | Zero-filled conditioning with the same shape as the positive conditioning, used as an empty negative reference. | CONDITIONING |

> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/Trellis2Conditioning/en.md)

---
**Source fingerprint (SHA-256):** `467698e58558ceca9ac633d63aacf360a1eb674ac4ebd47de7423f85e62c0fe6`
