Google API Pricing, Models & Rate Limits (2026)
Google serves the Gemini family through both a direct Gemini API and Vertex AI. Its distinguishing move is context window: Gemini 3.1 Pro's 2M-token window is the largest of any model on this site, and every current Gemini model reads native audio and video, not just text and images.
Also known as: Gemini, Google AI Studio, Vertex AI.
Models and prices
| Model | Input /M | Output /M | Blended /M |
|---|---|---|---|
| Gemini 3.5 Flash Lite | $0.25 | $1.50 | $0.56 |
| Gemini 3.6 Flash | $1.50 | $9.00 | $3.38 |
| Gemini 3.1 Pro | $2.00 | $12.00 | $4.50 |
5 legacy Google models
| Gemini 2.5 Flash Lite | $0.18/M blended |
| Gemini 3.1 Flash Lite | $0.56/M blended |
| Gemini 2.5 Flash | $0.85/M blended |
| Gemini 3.1 Flash | $1.69/M blended |
| Gemini 3.5 Flash | $3.38/M blended |
Speed
Fastest measured Google model is Gemini 3.5 Flash Lite at 162 tokens/sec (240ms TTFT), median across measured Google models is 114 tokens/sec. See the full speed benchmark methodology.
Operational facts
Verified 2026-08-08 against source.
| OpenAI-compatible | Partial |
| API base URL | https://generativelanguage.googleapis.com/v1beta |
| Auth model | API key (header or query param); OAuth/service-account on Vertex AI |
| Prompt caching | Yes |
| Batch discount | 50% |
| Free tier | Free tier with daily request cap on Google AI Studio |
| Rate-limit model | Free, then Tier 1-3, promoted by billing status |
| Data residency | Global by default; Vertex AI offers selectable regional endpoints |
| Trains on API data | No |
| SLA published | Yes |
Lifecycle
Google has 5 legacy models still routable. Full dates and successors on the model deprecation tracker.
Switching to and from Google
Calling Google through All AI Ask
Calling Google directly means adapting client code away from the plain OpenAI request shape — our gateway removes that: every model, including Google's, is called the same way.
curl https://allaiask.com/api/v1/chat \
-H "Authorization: Bearer $ALLAIASK_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gemini-3.5-flash-lite", "messages": [{"role": "user", "content": "Hello"}]}'FAQ
Is Google OpenAI-compatible?
Partially. Google publishes an OpenAI-compatible mode for some endpoints, but not a full 1:1 replacement for its native API — check https://ai.google.dev/gemini-api/docs before relying on it for every feature you use.
Does Google support prompt caching?
Yes, as of 2026-08-08 — see https://ai.google.dev/gemini-api/docs for the current mechanics and discount.
Does Google have a free tier?
Yes — Free tier with daily request cap on Google AI Studio.
How much does the Google API cost?
Current Google models range from $0.56 to $4.50 per million blended tokens (3:1 input:output). Full per-model pricing is in the table below.
Where is Google API data hosted?
Global by default; Vertex AI offers selectable regional endpoints