Download the PHP package muhammedsaidckr/gemini-client without Composer

Gemini PHP is a community-maintained PHP API client that allows you to interact with the Gemini AI API.

• Muhammed Said Çakır github.com/muhammedsaidckr

Table of Contents

• Prerequisites
• Setup
  • Installation
  • Setup your API key
• Usage
  • Chat Resource
    • Text-only Input
    • Text-and-image Input
    • Multi-turn Conversations (Chat)
    • Stream Generate Content
    • Count tokens
    • Configuration
  • Embedding Resource
  • Models
    • List Models
    • Get Model
• Troubleshooting
• Testing

Prerequisites

To complete this quickstart, make sure that your development environment meets the following requirements:

• Requires PHP 8.1+

Setup

Installation

First, install Gemini via the Composer package manager:

Ensure that the php-http/discovery composer plugin is allowed to run or install a client manually if your project does not already have a PSR-18 client integrated.

Setup your API key

To use the Gemini API, you'll need an API key. If you don't already have one, create a key in Google AI Studio.

Get an API key

Usage

Interact with Gemini's API:

If necessary, it is possible to configure and create a separate client.

Chat Resource

Text-only Input

Generate a response from the model given an input message. If the input contains only text, use the gemini-pro model.

Text-and-image Input

If the input contains both text and image, use the gemini-pro-vision model.

Multi-turn Conversations (Chat)

Using Gemini, you can build freeform conversations across multiple turns.

The gemini-pro-vision model (for text-and-image input) is not yet optimized for multi-turn conversations. Make sure to use gemini-pro and text-only input for chat use cases.

Stream Generate Content

By default, the model returns a response after completing the entire generation process. You can achieve faster interactions by not waiting for the entire result, and instead use streaming to handle partial results.

Count tokens

When using long prompts, it might be useful to count tokens before sending any content to the model.

Configuration

Every prompt you send to the model includes parameter values that control how the model generates a response. The model can generate different results for different parameter values. Learn more about model parameters.

Also, you can use safety settings to adjust the likelihood of getting responses that may be considered harmful. By default, safety settings block content with medium and/or high probability of being unsafe content across all dimensions. Learn more about safety settings.

Embedding Resource

Embedding is a technique used to represent information as a list of floating point numbers in an array. With Gemini, you can represent text (words, sentences, and blocks of text) in a vectorized form, making it easier to compare and contrast embeddings. For example, two texts that share a similar subject matter or sentiment should have similar embeddings, which can be identified through mathematical comparison techniques such as cosine similarity.

Use the embedding-001 model with either embedContents or batchEmbedContents:

Models

List Models

Use list models to see the available Gemini models:

Get Model

Get information about a model, such as version, display name, input token limit, etc.

Troubleshooting

Timeout

You may run into a timeout when sending requests to the API. The default timeout depends on the HTTP client used.

You can increase the timeout by configuring the HTTP client and passing in to the factory.

This example illustrates how to increase the timeout using Guzzle.

Testing

The package provides a fake implementation of the Gemini\Client class that allows you to fake the API responses.

To test your code ensure you swap the Gemini\Client class with the Gemini\Testing\ClientFake class in your test case.

The fake responses are returned in the order they are provided while creating the fake client.

All responses are having a fake() method that allows you to easily create a response object by only providing the parameters relevant for your test case.

In case of a streamed response you can optionally provide a resource holding the fake response data.

After the requests have been sent there are various methods to ensure that the expected requests were sent:

To write tests expecting the API request to fail you can provide a Throwable object as the response.