Gemini Connection

The Gemini connection type enables automations to call Google's Gemini API for multimodal language generation, summarisation, code assistance, and embedding tasks. It supports two access paths: Google AI Studio (simpler, API-key based, ideal for prototyping) and Vertex AI (enterprise-grade, GCP-integrated, recommended for production).

Connection Details

Property	Value
Type ID	GEMINI_AI
Category	AI Platforms
Testable	Yes

Friendly Example

What you enter	Example
Connection name	Gemini - Release Notes Drafter
Used for	Generating first-draft release notes from a list of merged pull request titles
Main details to collect	AI Studio API key (or Vertex AI service account JSON + GCP project ID)
Best person to provide it	AI platform owner or the GCP project administrator

After it is saved, workflow builders select Gemini - Release Notes Drafter in a node instead of re-entering credentials every time.

Mandatory Fields

The required fields depend on which access path you are using.

Google AI Studio (API Key)

Field	What It Means	Example
apiKey	The API key generated in Google AI Studio. Encrypted at rest.	AIzaSyAbc123...

Vertex AI (Service Account)

Field	What It Means	Example
serviceAccountJson	The full contents of the JSON key file downloaded from GCP. Encrypted at rest.	{ "type": "service_account", "project_id": "my-project", ... }
projectId	The GCP project ID where Vertex AI is enabled.	my-gcp-project-123

Which path should I use?

Use Google AI Studio for quick prototyping, internal tools, or when you do not have a GCP project. Use Vertex AI for production workloads, enterprise billing, private networking, or when you need data residency guarantees.

Advanced Fields

Field	What It Means	Example
defaultModel	The Gemini model to use when a workflow step does not specify one.	gemini-2.5-flash
region	The GCP region for Vertex AI inference. Ignored for AI Studio. Defaults to us-central1.	europe-west4
temperature	Controls output randomness (0.0 = deterministic, 1.0 = creative).	0.4
maxOutputTokens	Default maximum number of tokens the model may generate per response.	8192

Supported models (as of April 2026)

Model ID	Description
gemini-2.5-pro	Highest capability — complex reasoning, multimodal inputs, long context
gemini-2.5-flash	Balanced speed and intelligence — recommended for most production workflows
gemini-2.0-flash	Efficient and fast — best for high-volume, latency-sensitive steps

Rate limits vary by access path and billing tier. See Google's Gemini model documentation for up-to-date quotas.

Setup Instructions

Google AI Studio

1. Go to aistudio.google.com and sign in with a Google account.

2. Click "Get API key" in the left sidebar, then click Create API key in new project. Google creates a project and generates the key automatically.

3. Copy the API key. Paste it into Qinfinite's API Key field. The key can be retrieved again later from the same screen, unlike OpenAI or Anthropic.

4. (Optional) Enable billing. The free tier has strict quota limits. To increase quotas, link the underlying GCP project to a billing account via console.cloud.google.com → Billing.

Vertex AI

1. Open the GCP Console at console.cloud.google.com and select or create the project that will run Vertex AI workloads.

2. Enable the Vertex AI API. Go to APIs & Services → Library, search for Vertex AI API, and click Enable.

3. Create a Service Account. Go to IAM & Admin → Service Accounts → Create Service Account. Give it a name such as qinfinite-gemini.

4. Grant the required role. On the same screen (or via IAM → Grant Access), assign the Vertex AI User role (roles/aiplatform.user) to the service account.

5. Download the JSON key. In the Service Accounts list, click the account you just created, go to the Keys tab, click Add Key → Create new key, select JSON, and click Create. Paste the entire contents of the downloaded file into Qinfinite's Service Account JSON field.

Troubleshooting

Symptom	Likely Cause	Fix
400 API key not valid	The AI Studio API key is wrong or was copied with extra whitespace	Regenerate the key at aistudio.google.com and update the connection
403 PERMISSION_DENIED	The service account does not have the Vertex AI User role, or Vertex AI API is not enabled in the project	Verify the IAM role and confirm the API is enabled in the GCP console
429 RESOURCE_EXHAUSTED	You have exceeded the requests-per-minute or tokens-per-minute quota for your tier	Wait and retry with exponential back-off; request a quota increase via the GCP console
404 model not found	The defaultModel value is incorrect or not available in the selected region	Use one of the supported model IDs listed above; for Vertex AI, verify the model is available in the configured region