Every Gemini API discount, side by side
| Discount | What it saves | The catch |
|---|---|---|
| Free tier | Free input and output tokens | Some models only, low limits, and Google uses your content to improve its products |
| Batch API | 50% | Results within about 24 hours |
| Flex inference | 50% on ordinary requests | Preview, variable latency, best effort |
| Context caching | Repeated input at a fraction of the input price | Explicit caches also bill storage by the hour |
| A smaller model | Flash and Flash-Lite cost far less than Pro | Weaker on hard tasks |
| A lower price per token | 58–71% below list on Zurelay | OpenAI-compatible chat, not Google's Batch or Flex |
The free tier, and its catch
Google gives free input and output tokens on some Gemini models through the free tier, with lower rate limits and no access to context caching or the Batch API. The catch is data: on the free tier, Google uses what you send to improve its products. On the paid tier it doesn't, for most models. Keep anything private or customer-owned off the free tier. Setting up a key is in how to get a Gemini API key.
Batch API and Flex inference: half price
The Batch API takes a file of requests at 50% off and aims to return results within 24 hours. Flex inference brings the same 50% to ordinary, synchronous requests: set service_tier to flex. It's in preview, with variable latency and best-effort availability, so it suits background jobs rather than a user waiting on an answer.
Context caching
When requests repeat the same start, Google passes on the saving automatically, and an explicit cache lets you store a long document or system prompt and reuse it, billed at a fraction of the input price plus an hourly storage fee. More in prompt caching explained.
Pick the cheapest model that does the job
List prices per 1M tokens, with Zurelay's beside them:
| Model | Input | Output | On Zurelay (in / out) |
|---|---|---|---|
| Gemini 4 Argon | $2.00 | $10.00 | $0.80 / $4.00 |
| Gemini 3.1 Pro | $2.00 | $12.00 | $0.50 / $3.45 |
| Gemini 3.8 Flash | $0.75 | $3.75 | $0.22 / $1.10 |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 | $0.12 / $1.04 |
Is there a Gemini API coupon code?
No. Google's pricing page lists no coupon or promo codes for the Gemini API; the free tier is its way in.
Pay less for every token
Zurelay sells Gemini models below Google's list price (58–71% below) through an OpenAI-compatible endpoint, from $10 of prepaid credit and with no subscription. On Gemini 3.1 Pro it's below even Google's Batch price:
| Model, in / out per 1M | Google list | Google Batch | Zurelay |
|---|---|---|---|
| Gemini 3.1 Pro | $2.00 / $12.00 | $1.00 / $6.00 | $0.50 / $3.45 |
| Gemini 3.8 Flash | $0.75 / $3.75 | $0.375 / $1.875 | $0.22 / $1.10 |
Zurelay doesn't serve Google's Batch API, Flex tier or explicit caches; for those, go direct. Every Gemini model and price is on the Gemini API page.
Discounts read from Google's Gemini API pricing and Flex inference docs on November 15, 2026.