Google Gemini

Generate text and schema-constrained JSON with Gemini models, create embeddings and count tokens.

Use Gemini models to summarise, extract and classify business records and to build semantic search.

Build a flow with Google Gemini Provider API documentation

Source actions

Destination actions

Data types

Validation status

Implemented · fixture tested.

Implementation is checked against provider-shaped fixtures for every action, authentication failures, rate limits, pagination and duplicate safety. Provider sandbox and authorised account validation are recorded separately when credential-backed runs are completed.

What you can map

content

Generate content
Destination
Not retried automatically
Generates a reply with generateContent. Supply a JSON schema to receive schema-constrained JSON.

embedding batches

Embed texts in batch
Destination
Not retried automatically
Creates embeddings for several texts with batchEmbedContents; one record per text.

embeddings

Embed text
Destination
Not retried automatically
Creates one embedding with embedContent.

models

Get model
Source
Metadata for one Gemini model.
List models
Source
Gemini models available to the key with token limits and supported methods.

token count

Count tokens
Source
Counts the tokens a prompt uses for a model (countTokens).

Authentication

Gemini API key (x-goog-api-key header)

Plans and access

Gemini API keys from Google AI Studio on the free or paid tier. Free-tier usage may be used by Google to improve products and has lower rate limits; use a paid-tier project for business data.

Test environment

The free tier can be used for testing with reduced limits.

Rate limits

Per-project requests and tokens per minute and requests per day by model and tier; HTTP 429 when exceeded.

Provider events

Gemini API does not deliver webhooks. Nexra event triggers are not yet available; flows run on demand or on a schedule.

Before you build a flow

Uses the v1beta generateContent API. Streaming, tools, file and media inputs, caching and the Interactions API are not exposed.
Structured output uses generationConfig.responseMimeType with responseJsonSchema. Google's reference now also lists a newer responseFormat field; it is not used yet.
Embedding options are sent in embedContentConfig; the older top-level taskType, title and outputDimensionality fields are deprecated.
Token counting is a read; Google's reference does not state its price, so verify on the account's billing page.
Generation and embeddings are billed; they must be enabled per connection and are never retried automatically.