Gemini API key

Settings field: GEMINI_API_KEY · Expected format: AIza... or AQ...

Paste the key in Settings → BYOK, click Save keys, and wait for Connected (or Saved for keys that are not probed from the browser).

Used for

Summarizes news into catalyst labels when RSS headlines lack dates, powers Catalyst Agent™, and optional Google Search grounding on thin excerpts. Uses gemini-3.1-flash-lite.

Pricing

Gemini API has a free tier in Google AI Studio plus pay-as-you-go billing on the Paid tier. Kvantrank uses gemini-3.1-flash-lite for catalyst labeling.

• Free tier: $0 (Limited free input and output tokens in AI Studio)
• gemini-3.1-flash-lite (Paid): $0.25/M input, $1.50/M output (Model Kvantrank uses for catalyst labeling)
• Typical Kvantrank usage: About $0.05/day (Roughly 1,000 tokens per daily pipeline run on average)
• Google Search grounding: $14 / 1,000 queries (After 5,000 free grounded prompts/month - paid tier only)

Official pricing: https://ai.google.dev/gemini-api/docs/pricing

Batch API and context caching offer lower per-token rates. Enable billing in Google AI Studio to move from Free to Paid. Grounding adds cost when used.

Sign up

https://aistudio.google.com/apikey

How to get the key

1. Sign in with your Google account at Google AI Studio.
2. Click Create API key and copy the key.
3. Paste it into Settings → BYOK and save.

Notes

Settings shows Saved rather than Connected because the key is validated at pipeline runtime. Model: gemini-3.1-flash-lite. Typical daily use is about 1,000 tokens (~$0.05/day on Paid tier rates). Google Search grounding, when used, is billable per search query.