gemini-go-workshop

Workshop to discover Gemini 3.5 using the Go SDK

Google AI API key

Create an API key at https://aistudio.google.com/apikey

Google Cloud project

Create a project at https://console.cloud.google.com/

Prerequistes

Go 1.23+

Type go version to make sure.

For samples requiring API key

export GOOGLE_GENAI_USE_VERTEXAI=false
export GOOGLE_API_KEY=<YOUR_API_KEY>

For samples requiring Google Cloud Vertex AI

export GOOGLE_GENAI_USE_VERTEXAI=true
export GOOGLE_CLOUD_PROJECT=deleplace-ai-demos
export GOOGLE_CLOUD_LOCATION=us-central1
gcloud auth application-default login
gcloud services enable aiplatform.googleapis.com

Read the samples

Open in your editor the source files sample*.go

Run the samples

Sample 0: Text input

go run . -n=0

Uncomment the code section to discover the structured response from the model.

Exercise: ask the same question but in French.

Sample 1: Streaming text output

go run . -n=1

Sample 2: Multimodal image input

go run . -n=2

Exercise: Is the last answer good enough? What happens if you use the model "gemini-3.1-pro-preview" instead?

Sample 3: Multimodal audio input

go run . -n=3

Sample 4: Multimodal video input

go run . -n=4

Exercise: instead of a text question, provide the audio file ./testdata/question_about_video.mp3 as the question.

Sample 5: Image generation

go run . -n=5

Exercise: write an extremely specific prompt to generate an image.

Note that safety filters are very sensitive and often refuse harmless prompts.

Sample 6: Image upscaling

go run . -n=6