README

This is a PHP library for Ollama. Ollama is an open-source project that serves as a powerful and user-friendly platform for running LLMs on your local machine. It acts as a bridge between the complexities of LLM technology and the desire for an accessible and customizable AI experience.

☕️ Buy me a coffee

Whether you use this project, have learned something from it, or just like it, please consider supporting it by buying me a coffee, so I can dedicate more time on open-source projects like this :)

Table of Contents

• Get Started
• Usage
  • Completions Resource
    • create()
    • createStreamed()
  • Chat Resource
    • create()
    • createStreamed()
  • Models Resource
    • list()
    • show()
    • create()
    • createStreamed()
    • copy()
    • delete()
    • pull()
    • pullStreamed()
    • push()
    • pushStreamed()
    • runningList()
  • Blobs Resource
    • exists()
    • create()
  • Embed Resource
    • create()
• Testing
• Changelog
• Contributing
• Credits
• License

Get Started

Requires PHP 8.1+
Requires Ollama

You can find Official Ollama documentation here.

First, install Ollama PHP via the Composer package manager:

composer require ardagnsrn/ollama-php

Then, you can create a new Ollama client instance:

// with default base URL
$client = \ArdaGnsrn\Ollama\Ollama::client();

// or with custom base URL
$client = \ArdaGnsrn\Ollama\Ollama::client('http://localhost:11434');

Usage

Completions Resource

create

Generate a response for a given prompt with a provided model.

$completions = $client->completions()->create([
    'model' => 'llama3.1',
    'prompt' => 'Once upon a time',
]);

$completions->response; // '...in a land far, far away...'

$response->toArray(); // ['model' => 'llama3.1', 'response' => '...in a land far, far away...', ...]

createStreamed

Generate a response for a given prompt with a provided model and stream the response.

$completions = $client->completions()->createStreamed([
    'model' => 'llama3.1',
    'prompt' => 'Once upon a time',
]);


foreach ($completions as $completion) {
    echo $completion->response;
}
// 1. Iteration: '...in'
// 2. Iteration: ' a'
// 3. Iteration: ' land'
// 4. Iteration: ' far,'
// ...

Chat Resource

create

Generate a response for a given prompt with a provided model.

$response = $client->chat()->create([
    'model' => 'llama3.1',
    'messages' => [
        ['role' => 'system', 'content' => 'You are a llama.'],
        ['role' => 'user', 'content' => 'Hello!'],
        ['role' => 'assistant', 'content' => 'Hi! How can I help you today?'],
        ['role' => 'user', 'content' => 'I need help with my taxes.'],
    ],
]);

$response->message->content; // 'Ah, taxes... *chew chew* Hmm, not really sure how to help with that.'

$response->toArray(); // ['model' => 'llama3.1', 'message' => ['role' => 'assistant', 'content' => 'Ah, taxes...'], ...]

Also, you can use the tools parameter to provide custom functions to the chat. tools parameter can not be used with createStreamed method.

$response = $client->chat()->create([
    'model' => 'llama3.1',
    'messages' => [
        ['role' => 'user', 'content' => 'What is the weather today in Paris?'],
    ],
    'tools' => [
        [
            'type' => 'function',
            'function' => [
                'name' => 'get_current_weather',
                'description' => 'Get the current weather',
                'parameters' => [
                    'type' => 'object',
                    'properties' => [
                        'location' => [
                            'type' => 'string',
                            'description' => 'The location to get the weather for, e.g. San Francisco, CA',
                        ],
                        'format' => [
                            'type' => 'string',
                            'description' => 'The location to get the weather for, e.g. San Francisco, CA',
                            'enum' => ['celsius', 'fahrenheit']
                        ],
                    ],
                    'required' => ['location', 'format'],
                ],
            ],
        ]
    ]
]);

$toolCall = $response->message->toolCalls[0];

$toolCall->function->name; // 'get_current_weather'
$toolCall->function->arguments; // ['location' => 'Paris', 'format' => 'celsius']

$response->toArray(); // ['model' => 'llama3.1', 'message' => ['role' => 'assistant', 'toolCalls' => [...]], ...]

createStreamed

Generate a response for a given prompt with a provided model and stream the response.

$responses = $client->chat()->createStreamed([
    'model' => 'llama3.1',
    'messages' => [
        ['role' => 'system', 'content' => 'You are a llama.'],
        ['role' => 'user', 'content' => 'Hello!'],
        ['role' => 'assistant', 'content' => 'Hi! How can I help you today?'],
        ['role' => 'user', 'content' => 'I need help with my taxes.'],
    ],
]);


foreach ($responses as $response) {
    echo $response->message->content;
}
// 1. Iteration: 'Ah,'
// 2. Iteration: ' taxes'
// 3. Iteration: '... '
// 4. Iteration: ' *chew,'
// ...

Models Resource

list

List all available models.

$response = $client->models()->list();

$response->toArray(); // ['models' => [['name' => 'llama3.1', ...], ['name' => 'llama3.1:80b', ...], ...]]

show

Show details of a specific model.

$response = $client->models()->show('llama3.1');

$response->toArray(); // ['modelfile' => '...', 'parameters' => '...', 'template' => '...']

create

Create a new model.

$response = $client->models()->create([
    'name' => 'mario',
    'modelfile' => "FROM llama3.1\nSYSTEM You are mario from Super Mario Bros."
]);

$response->status; // 'success'

createStreamed

Create a new model and stream the response.

$responses = $client->models()->createStreamed([
    'name' => 'mario',
    'modelfile' => "FROM llama3.1\nSYSTEM You are mario from Super Mario Bros."
]);

foreach ($responses as $response) {
    echo $response->status;
}

copy

Copy an existing model.

$client->models()->copy('llama3.1', 'llama3.2'); // bool

delete

Delete a model.

$client->models()->delete('mario'); // bool

pull

Pull a model from the Ollama server.

$response = $client->models()->pull('llama3.1'); 
$response->toArray() // ['status' => 'downloading digestname', 'digest' => 'digestname', 'total' => 2142590208, 'completed' => 241970]

pullStreamed

Pull a model from the Ollama server and stream the response.

$responses = $client->models()->pullStreamed('llama3.1'); 

foreach ($responses as $response) {
    echo $response->status; 
}

push

Push a model to the Ollama server.

$response = $client->models()->push('llama3.1');
$response->toArray() // ['status' => 'uploading digestname', 'digest' => 'digestname', 'total' => 2142590208]

pushStreamed

Push a model to the Ollama server and stream the response.

$responses = $client->models()->pushStreamed('llama3.1');

foreach ($responses as $response) {
    echo $response->status; 
}

runningList

List all running models.

$response = $client->models()->runningList();

$response->toArray(); // ['models' => [['name' => 'llama3.1', ...], ['name' => 'llama3.1:80b', ...], ...]]

Blobs Resource

exists

Check if a blob exists.

$client->blobs()->exists('blobname'); // bool

create

Create a new blob.

$client->blobs()->create('blobname'); // bool

Embed Resource

create

Generate an embedding for a given text with a provided model.

$response = $client->embed()->create([
    'model' => 'llama3.1',
    'input' => [
        "Why is the sky blue?",
    ]
]);

$response->toArray(); // ['model' => 'llama3.1', 'embedding' => [0.1, 0.2, ...], ...]

Testing

composer test

Changelog

Please see CHANGELOG for more information on what has changed recently.

Contributing

Please see CONTRIBUTING for details.

Credits

• Arda GUNSUREN
• All Contributors

License

The MIT License (MIT). Please see License File for more information.