# Image to embeddings
> Supported in: Batch
Converts images into embeddings using the provided model.
**Expression categories:** Media
## Declared arguments
* **Media reference:** Column containing media references (images) to convert to embeddings.
*Expression\*
* **Model:** The embedding model to use for the conversion. Must support vision capabilities.
*Model*
* *optional* **Output mode:** Choose between simply returning the output or returning a struct, containing either the output in a field named 'ok', or any errors in a field named 'errors'.
*Enum\*
**Output type:** *Embedded vector*
## Examples
### Example 1: Base case
**Description:** Example embeddings for an image.
**Argument values:**
* **Media reference:** `mediaRef`
* **Model:**
googleSiglip2Embedding(
)
* **Output mode:** *null*
| mediaRef | **Output** |
| ----- | ----- |
| {
"mimeType": "image/jpeg",
"reference": {
"type": "mediaSetViewItem",
"... | embeddings-result |
***
### Example 2: Null case
**Description:** Null input should have a null output.
**Argument values:**
* **Media reference:** `mediaRef`
* **Model:**
googleSiglip2Embedding(
)
* **Output mode:** *null*
| mediaRef | **Output** |
| ----- | ----- |
| *null* | *null* |
***