Build AI agents on Google Gemini in YAML

Gemini's headline strengths are context window size (up to 2M tokens) and multimodal handling. For agents that need to reason over long documents, video, or audio, Gemini removes constraints other providers still impose. Pricing on Flash is competitive for high-volume tasks.

Models worth knowing about

Drop into your app.yaml

1brain:2  provider: google3  model: gemini-2.0-pro4  config:5    api_key: "{{env.GOOGLE_API_KEY}}"6  temperature: 0.27  max_tokens: 8192

Google Geminidoesn't have a first-class catalog entry yet. Configure it inline using env templates, the same way the blog examples show. Native catalog support is on the roadmap.

# add to ~/.digitorn/.env
GOOGLE_API_KEY=...

Install Digitorn and chat with your Google Gemini agent

# 1. install the runtime
curl -sSL https://digitorn.ai/install | sh

# 2. drop your Google Gemini key in the env file
echo 'GOOGLE_API_KEY=...' >> ~/.digitorn/.env

# 3. install a starter agent and chat
digitorn install hub://digitorn/digitorn-code
digitorn chat digitorn-code