---
title: "Gemma API key and pricing"
canonical: https://relayswitch.io/pricing/gemma
generated: 2026-11-02T12:08:19.053Z
source: https://api.relayswitch.io/v1/catalog
base_url: https://api.relayswitch.io/v1
---

Checked 2026-10-26

# Gemma API key and pricing

Gemma input, cached-input and output prices on one key, read live from the catalog. Prepaid by card or crypto, no KYC, no subscription.

Gemma 4 31B for the hardest prompts, Gemma 4 26B-A4B for most work at a lower rate, the 3n ids for short high-volume calls. Open weights, so the same model runs locally; here it runs on a prepaid key with no GPU. Every row below is a live catalog read; prices are USD per million tokens and the developer list appears where recorded.

## 01Every Gemma model

| Model | Input | Cached | Output | Developer list in / out | Context |
| --- | --- | --- | --- | --- | --- |
| [Gemma 3 27B](https://relayswitch.io/models/gemma-3-27b-it) | $1.20 | $0.80 | $1.20 | — | 131K tokens |
| [Gemma 3n E2B](https://relayswitch.io/models/gemma-3n-e2b-it) | $0.12 | input rate | $0.08 | — | 8K tokens |
| [Gemma 3n E4B](https://relayswitch.io/models/gemma-3n-e4b-it) | $0.20 | input rate | $0.16 | — | 8K tokens |
| [Gemma 4 12B](https://relayswitch.io/models/gemma-4-12b-it) | $0.40 | $0.20 | $1.20 | — | — |
| [Gemma 4 26B-A4B](https://relayswitch.io/models/gemma-4-26b-a4b-it) | $0.56 | $0.3055 | $1.60 | — | — |
| [Gemma 4 31B](https://relayswitch.io/models/gemma-4-31b-it) | $3.00 | $2.4546 | $4.00 | — | — |

USD per million tokens, read from the RelaySwitch catalog when this page was built. Developer list: the catalog's record of the developer's own price.

## 02Which model should I use?

Gemma 4 31B for the hardest prompts, Gemma 4 26B-A4B for most work at a lower rate, the 3n ids for short high-volume calls. Open weights, so the same model runs locally; here it runs on a prepaid key with no GPU. Compare the worked models below before choosing for production.

## 03Worked models

-   [Gemma 4 31B](https://relayswitch.io/models/gemma-4-31b-it) — 3.0000 input / 4.0000 output per million tokens.
-   [Gemma 4 26B-A4B](https://relayswitch.io/models/gemma-4-26b-a4b-it) — 0.5600 input / 1.6000 output per million tokens.
-   [Gemma 3 27B](https://relayswitch.io/models/gemma-3-27b-it) — 1.2000 input / 1.2000 output per million tokens.

## Questions

Is Gemma billed monthly?

No. Credit is prepaid and each request bills per token.

Can I pay for Gemma with crypto?

Yes: USDT, USDC, BTC, ETH or SOL, alongside card.

Is the Gemma AI API free?

The weights are free to download and run yourself. A hosted API bills per token; here Gemma runs on a prepaid key with no monthly fee, at the live rates in the table.

Is Gemma 4 available via API?

Yes. gemma-4-31b-it, gemma-4-26b-a4b-it and gemma-4-12b-it take the OpenAI chat format on this key; send the id as the model.

How much does the Gemma 4 API cost?

The table above is read from the catalog at render: input and output per million tokens for every Gemma id, no minimum, no subscription.

What is Gemma AI good for?

Chat, extraction, classification and summarisation where a small open model is enough and price per token matters. The 31B id handles longer reasoning; the 3n ids answer fast and cheap.