- •The Generate content action generates a response from the model based on the given input. It creates a requestBody that contains fields with different types, such as string or array.
You must insert the access_token field from the Get access token action and the model_ID field to call the model, and create a payload. For example, the model_ID value can be gemini-1.5-flash-001 or gemini-1.5-pro-001.
The following snippet is a sample to build a requestBody object:
If fields in the requestBody need to be in different formats, you must add additional namespaces as shown in the following sample code:
- •The Function calling action generates a response from the model based on the given input with the Function Calling API.
You must insert the access_token field from the Get access token action and the model_ID field to call the model, and create a payload. For example, the model_ID value can be gemini-1.5-flash-001 or gemini-1.5-pro-001.
The following snippet is a sample to build a requestBody object:
If fields in the requestBody need to be in different formats, you must add additional namespaces as shown in the following sample code:
- •The Grounding action generates a response from the model based on the given input and connects the model output to verifiable sources of information.
You must insert the access_token field from the Get access token action and the model_ID field to call the model, and create a payload. For example, the model_ID value can be gemini-1.5-flash-001 or gemini-1.5-pro-001.
The following snippet is a sample to build a requestBody object:
If fields in the requestBody need to be in different formats, you must add additional namespaces as shown in the following sample code:
- •The Text embeddings action generates an embedding from the model based on the given input.
You must insert the access_token field from the Get access token action and the model_ID field to call the model, and create a payload. For example, the model_ID value can be textembedding-gecko-multilingual@001 or text-multilingual-embedding-002.
The following snippet is a sample to build a requestBody object:
<root>
<instances>
<task_type>RETRIEVAL_DOCUMENT</task_type>
<title>document title</title>
<content>I would like embeddings for this text!</content>
</instances>
</root>
If fields in the requestBody need to be in different formats, you must add additional namespaces as shown in the following sample code:
- •The Multimodal embeddings action generates an embedding from the model based on the given input, which can include a combination of image, text, and video data.
You must insert the access_token field from the Get access token action and the model_ID field to call the model, and create a payload. For example, the model_ID value can be multimodalembedding@001.
The following snippet is a sample to build a requestBody object:
If fields in the requestBody need to be in different formats, you must add additional namespaces as shown in the following sample code: