# Google Gemini

> Direct access to Google Gemini models through the gemini API.

Source: https://tokenmojo.ai/providers/google-gemini

## Stats

- Models published: 7
- Average quality: 59.5
- Median speed: 4.6 tok/s
- Average reliability: 83.5%
- Total samples: 1123

## Description

This integration connects to the official gemini API for Google Gemini inference. It enables factual comparison of latency, pricing, and performance metrics.

## Models

- [Gemini 3.5 Flash Lite](https://tokenmojo.ai/models/google-gemini/gemini-3-5-flash-lite)
- [Gemini 3.6 Flash](https://tokenmojo.ai/models/google-gemini/gemini-3-6-flash)
- [Gemini Flash Latest](https://tokenmojo.ai/models/google-gemini/gemini-flash-latest): Google Gemini Flash Latest provides 1M token context and reasoning capabilities.
- [Gemini Omni Flash Preview](https://tokenmojo.ai/models/google-gemini/gemini-omni-flash-preview)
- [Gemini Pro Latest](https://tokenmojo.ai/models/google-gemini/gemini-pro-latest): Google Gemini Pro Latest provides a 1M token context window and reasoning capabilities for scale.
- [Gemma 4 26B A4B IT](https://tokenmojo.ai/models/google-gemini/gemma-4-26b-a4b-it): Google Gemma 4 26B IT: 262K Context, 32K Max Output, Reasoning Capabilities
- [Gemma 4 31B IT](https://tokenmojo.ai/models/google-gemini/gemma-4-31b-it): Google Gemini Gemma 4 31B IT: 262K context, 32K output, reasoning enabled.