> For the complete documentation index, see [llms.txt](https://docs.roboflow.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.roboflow.com/workflows/blocks/blocks/run-a-model/embedding-model.md).

# Embedding Model

Generate image embeddings using a classification model.

Extract raw features before the final linear classifier or logits before Softmax/Sigmoid of a workspace ResNet, ViT or DINOv3 model, or a pretrained ResNet. Uses the model's original preprocessing. Compare vectors only from the same embedding space.

### Type identifier

Use the following identifier in step `"type"` field: `roboflow_core/embedding_model@v1` to add the block as a step in your workflow.

### Properties

| **Name**      | **Type** | **Description**                                                                                        | Refs |
| ------------- | -------- | ------------------------------------------------------------------------------------------------------ | ---- |
| `name`        | `str`    | Enter a unique identifier for this step..                                                              | ❌    |
| `model_id`    | `str`    | Roboflow model identifier..                                                                            | ✅    |
| `output_type` | `str`    | Select Feature Vector (input to the final linear layer) or Logits (output before Softmax or Sigmoid).. | ❌    |

The **Refs** column marks possibility to parametrise the property with dynamic values available in `workflow` runtime. See *Bindings* for more info.

### Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds `Embedding Model` in version `v1` has.

<details>

<summary>Input and output bindings</summary>

* input
  * `data` ([*`image`*](https://docs.roboflow.com/workflows/developer-guide/developer-guide/kinds/image.md)): Image or crop to embed..
  * `model_id` ([*`roboflow_model_id`*](https://docs.roboflow.com/workflows/developer-guide/developer-guide/kinds/roboflow-model-id.md)): Roboflow model identifier..
* output
  * `embedding` ([`embedding`](https://docs.roboflow.com/workflows/developer-guide/developer-guide/kinds/embedding.md)): A list of floating point numbers representing a vector embedding..
  * `embedding_info` ([`dictionary`](https://docs.roboflow.com/workflows/developer-guide/developer-guide/kinds/dictionary.md)): Dictionary.

</details>

<details>

<summary>Example JSON definition</summary>

```json
{
	    "name": "<your_step_name_here>",
	    "type": "roboflow_core/embedding_model@v1",
	    "data": "$inputs.image",
	    "model_id": "my_project/3",
	    "output_type": "feature_vector"
	}
```

</details>
