Skip to main content

​
OpenRouter Provider

The OpenRouter provider gives you unified access to models from multiple providers including Google, Mistral, Meta, Microsoft, and others through a single API.

​
Configuration

​
API Key Setup

Set your OpenRouter API key in your .env file: 
OPENROUTER_API_KEY=your_api_key_here
The API key configuration is defined in config/llm-magic.php: 
'apis' => [
    'openrouter' => [
        'token' => env('OPENROUTER_API_KEY'),
    ],
]

​
Available Models

LLM Magic supports the following models through OpenRouter:

​
xAI Models

  • x-ai/grok-beta - xAI’s Grok model

​
Google Gemini Models

  • google/gemini-pro-1.5 - Gemini Pro 1.5
  • google/gemini-flash-1.5-8b - Gemini Flash 1.5 8B
  • google/gemini-flash-1.5 - Gemini Flash 1.5

​
Mistral Models

  • mistralai/mistral-7b-instruct - Mistral 7B Instruct
  • mistralai/mistral-large - Mistral Large
  • mistralai/mistral-nemo - Mistral Nemo
  • mistralai/mistral-tiny - Mistral Tiny

​
Cohere Models

  • cohere/command-r-08-2024 - Cohere Command R

​
Meta Llama Models

  • meta-llama/llama-3.1-8b-instruct - Llama 3.1 8B Instruct
  • meta-llama/llama-3.1-70b-instruct - Llama 3.1 70B Instruct
  • meta-llama/llama-3.1-405b-instruct - Llama 3.1 405B Instruct
  • meta-llama/llama-3.1-405b-instruct:free - Llama 3.1 405B Instruct (Free)
  • meta-llama/llama-3.2-1b-instruct - Llama 3.2 1B Instruct
  • meta-llama/llama-3.2-3b-instruct - Llama 3.2 3B Instruct
  • meta-llama/llama-3.2-11b-vision-instruct - Llama 3.2 11B Vision
  • meta-llama/llama-3.2-90b-vision-instruct - Llama 3.2 90B Vision

​
Microsoft Models

  • microsoft/wizardlm-2-8x22b - WizardLM 2 8x22B

​
Nous Research Models

  • nousresearch/hermes-3-llama-3.1-405b - Hermes 3 Llama 3.1 405B
  • nousresearch/hermes-3-llama-3.1-405b:free - Hermes 3 Llama 3.1 405B (Free)

​
Perplexity Models

  • perplexity/llama-3.1-sonar-large-128k-chat - Llama 3.1 Sonar 70B
  • perplexity/llama-3.1-sonar-large-128k-online - Llama 3.1 Sonar 70B (Online)

​
Model Constants

Use these constants for type safety: 
use Mateffy\Magic\Models\OpenRouter;

OpenRouter::X_AI_GROK_BETA
OpenRouter::GOOGLE_GEMINI_1_5_PRO
OpenRouter::GOOGLE_GEMINI_1_5_FLASH_8B
OpenRouter::GOOGLE_GEMINI_1_5_FLASH
OpenRouter::MISTRAL_7B_INSTRUCT
OpenRouter::MISTRAL_LARGE
OpenRouter::MISTRAL_NEMO
OpenRouter::MISTRAL_TINY
OpenRouter::COHERE_COMMAND_R
OpenRouter::META_LLAMA_3_1_8B_INSTRUCT
OpenRouter::META_LLAMA_3_1_70B_INSTRUCT
OpenRouter::META_LLAMA_3_1_405B_INSTRUCT
OpenRouter::META_LLAMA_3_1_405B_INSTRUCT_FREE
OpenRouter::META_LLAMA_3_2_1B_INSTRUCT
OpenRouter::META_LLAMA_3_2_3B_INSTRUCT
OpenRouter::META_LLAMA_3_2_11B_VISION_INSTRUCT
OpenRouter::META_LLAMA_3_2_90B_VISION_INSTRUCT
OpenRouter::MICROSOFT_WIZARDLM_2_8X22B
OpenRouter::NOUS_HERMES_3_405B_INSTRUCT_FREE
OpenRouter::NOUS_HERMES_3_405B_INSTRUCT
OpenRouter::PERPLEXITY_LLAMA_3_1_SONAR_70B
OpenRouter::PERPLEXITY_LLAMA_3_1_SONAR_70B_ONLINE

​
Usage

​
Using the Constructor

Create an OpenRouter model instance: 
use Mateffy\Magic\Models\OpenRouter;
use Mateffy\Magic\Models\Options\ElElEmOptions;

$model = new OpenRouter(
    model: OpenRouter::GOOGLE_GEMINI_1_5_FLASH,
    options: new ElElEmOptions
);

​
Using Static Factory Methods

LLM Magic provides convenient static methods: 
use Mateffy\Magic\Models\OpenRouter;

// Use any model string
$model = OpenRouter::model('anthropic/claude-3-5-sonnet');

// Use xAI Grok
$model = OpenRouter::grok();

​
Getting Available Models

Retrieve a list of all available OpenRouter models: 
use Mateffy\Magic\Models\OpenRouter;

$models = OpenRouter::models();
// Returns a Collection with prefixed model names like 'openrouter/google/gemini-flash-1.5'

$models = OpenRouter::models(prefix: null, prefixLabels: null);
// Returns models without prefix

​
API Configuration

OpenRouter uses the OpenAI-compatible API with a custom base URI: 
protected function getOpenAiBaseUri(): ?string
{
    return 'openrouter.ai/api/v1';
}

​
Organization Info

The OpenRouter provider includes organization metadata:
  • ID: openrouter
  • Name: OpenRouter
  • Website: https://openrouter.ai
  • Privacy: Data may be used for model training and abuse prevention

​
Benefits of OpenRouter

​
Unified Access

Access models from multiple providers with a single API key:
  • No need to manage separate API keys for each provider
  • Consistent API interface across all models
  • Easy model switching and comparison

​
Cost Optimization

OpenRouter offers competitive pricing and free tier options:
  • Free models available (look for :free suffix)
  • Pay-as-you-go pricing
  • No monthly commitments

​
Model Discovery

Explore and experiment with models from different providers:
  • Test different models without setup overhead
  • Compare performance across providers
  • Find the best model for your use case

​
Advanced Usage

​
Using Any Model

OpenRouter supports many more models than those listed above. Use the model() static method to access any model: 
use Mateffy\Magic\Models\OpenRouter;

// Use any model available on OpenRouter
$model = OpenRouter::model('anthropic/claude-3-5-sonnet');
$model = OpenRouter::model('openai/gpt-4-turbo');
$model = OpenRouter::model('meta-llama/llama-3.3-70b-instruct');

​
Model Options

Customize model behavior with options: 
use Mateffy\Magic\Models\OpenRouter;
use Mateffy\Magic\Models\Options\ElElEmOptions;

$options = new ElElEmOptions(
    // Configure temperature, max tokens, etc.
);

$model = new OpenRouter(OpenRouter::GOOGLE_GEMINI_1_5_FLASH, $options);

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us