Google Gemini

Google Gemini offers free API access with a reasonable quota, very large context windows (up to 2M tokens on Pro), and strong instruction-following. It's the other great free starting point alongside Groq.

Get a key

Go to aistudio.google.com/apikey
Sign in with your Google account
Click Create API key in new project (or select an existing project)
Copy the key

Models available in ContentAI

Model	Context	Notes
`gemini-2.5-flash`	1M	⭐ Default — best quality/speed
`gemini-2.0-flash-exp`	1M	Experimental, fastest
`gemini-2.5-pro`	2M	Best reasoning, slowest
`gemini-1.5-flash`	1M	Older, cheaper, still good

Rate limits and quotas

Google's free tier has per-minute and per-day limits that vary by model. At time of writing:

gemini-2.5-flash — ~15 RPM, 1,500 requests/day
gemini-2.5-pro — ~2 RPM, 50 requests/day (paid tier recommended)

If you hit a limit, you'll see a 429 toast. Wait 60 seconds or switch to Groq.

Free tier may train on prompts

Google's free Gemini tier reserves the right to train on user inputs. If you're generating content for clients or products that should stay private, use the paid tier or switch to OpenAI/Anthropic/Groq.

Quality notes

Excellent at long-form. Gemini 2.5 Flash handles 3,000+ word blog posts gracefully.
Follows structure well. Great for templated outputs like listicles and outlines.
Occasionally refuses benign prompts. If you hit a safety refusal, rephrase or switch provider.

Setup in ContentAI

Settings → Google card → paste key
Pick gemini-2.5-flash
Click Test
(Optional) Set as active provider

Common errors

Error	Fix
`API key not valid. Please pass a valid API key.`	Generate a new key, make sure you copied it fully
`Resource has been exhausted (e.g. check quota).`	Free tier limit. Wait or switch model/provider
`Model not found`	The model ID is wrong — update `lib/store.ts`
`User location is not supported`	Gemini isn't available in your country. Use Groq/OpenAI

Next: OpenAI →