Connection

Credentials

List of scopes separated by spaces. If not defined, using default scope `https://www.googleapis.com/auth/cloud-platform` (see Google OAuth2 documentation for the list of scopes).

Rotate refresh tokens

Whether DSS should expect refresh tokens to be renewed when an access token is retrieved

Google Cloud region

Google Cloud project

Custom properties

Advanced connection properties

For specific use cases

Text completion models

More information about the pricing: Vertex AI Pricing

Gemini 2.5 Flash

Input: $0.00030/1k tokens - Output: $0.0025/1k tokens - Vertex documentation

Gemini 2.5 Flash Lite

Input: $0.00010/1k tokens - Output: $0.0004/1k tokens ( =< 200K input tokens) - Vertex documentation

Gemini 2.5 Pro

Input: $0.00125/1k tokens - Output: $0.01/1k tokens ( =< 200K input tokens) - Vertex documentation

Gemini 2.0 Flash

Fine-tuned for conversation - Input: $0.00015/1k tokens - Output: $0.0006/1k tokens

Gemini 2.0 Flash-Lite

Fine-tuned for conversation - Input: $0.000075/1k tokens - Output: $0.0003/1k tokens

Gemini 1.5 Flash

Input:$0.00001875/1k characters - Output: $0.000075/1k characters ( =< 128K context window)
Deprecated / scheduled for retirement on Sept 24th, 2025

Gemini 1.5 Pro

Input:$0.0003125/1k characters - Output: $0.00125/1k characters ( =< 128K context window)
Deprecated / scheduled for retirement on Sept 24th, 2025

Gemini 2.0 Flash (experimental)

Experimental preview - Vertex documentation
Depreciated, use non-experimental 2.0 Flash model instead.

Embedding models

More information about the pricing: Vertex AI Pricing

Gemini Text embedding

Embedding cost: $0.00015/1k characters

Text embedding (update 005)

Embedding cost: $0.000025/1k characters

Text embedding (update 004)

Embedding cost: $0.000025/1k characters

Text embedding (multilingual)

Embedding cost: $0.000025/1k characters

Multimodal embedding

Embedding cost: $0.0002/1k characters - $0.0001/image

Image generation models

Imagen 3

See Google documentation for costs

Imagen 3 Fast

See Google documentation for costs

Custom models

Model id

Required. A model ID supported by Vertex.

Model display name

Used to identify this model in DSS. If empty defaults to the model id.

Model type

Prompt cost

In $/1k tokens

Completion cost

In $/1k tokens

Embedding cost

In $/1k tokens

Max tokens

Embedding size