PHPackages laravel-local-llm/sdk - PHPackages - PHPackages [Skip to content](#main-content)[PHPackages](/)[Directory](/)[Categories](/categories)[Trending](/trending)[Leaderboard](/leaderboard)[Changelog](/changelog)[Analyze](/analyze)[Collections](/collections)[Log in](/login)[Sign up](/register) 1. [Directory](/) 2. / 3. [Utility & Helpers](/categories/utility) 4. / 5. laravel-local-llm/sdk ActiveLibrary[Utility & Helpers](/categories/utility) laravel-local-llm/sdk ===================== Modern Multi-Driver Local LLM Integration for Laravel v1.1.0(2mo ago)00MITPHPPHP ^8.2CI passing Since Mar 8Pushed 2mo agoCompare [ Source](https://github.com/shamimlaravel/Laravel-Local-LLM-SDK)[ Packagist](https://packagist.org/packages/laravel-local-llm/sdk)[ RSS](/packages/laravel-local-llm-sdk/feed)WikiDiscussions main Synced 1mo ago READMEChangelogDependencies (6)Versions (3)Used By (0) Laravel Local LLM SDK ===================== [](#laravel-local-llm-sdk) Modern, multi-driver, failover-ready local LLM integration for Laravel. [![PHP Version](https://camo.githubusercontent.com/a36c0210404104eacc59eff100cd75e4f7698b14b271d9870ac9331d7f7314ef/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f5048502d382e342532422d3737374242343f7374796c653d666c6174266c6f676f3d706870)](https://www.php.net/)[![Laravel Version](https://camo.githubusercontent.com/bd5c3a1f6b2800a6886739143d9019f2ba8e7b9174750a7669941d8c43e5f549/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f4c61726176656c2d31322532422d4646324432303f7374796c653d666c6174266c6f676f3d6c61726176656c)](https://laravel.com/)[![MIT License](https://camo.githubusercontent.com/784362b26e4b3546254f1893e778ba64616e362bd6ac791991d2c9e880a3a64e/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f4c6963656e73652d4d49542d677265656e2e737667)](LICENSE)[![Static Analysis](https://camo.githubusercontent.com/942bdbddc7b2adea1d63ed80793492d06d72ef41911edcba33310d0745581548/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f5048505374616e2d4c6576656c253230392d627269676874677265656e)](phpstan.neon)[![Code Style](https://camo.githubusercontent.com/8dce636d5374d5aed75936dd650f85479403f843643da105a5f1821d4d85f7f1/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f436f64652532305374796c652d50696e742d626c7565)](pint.json) Why Laravel Local LLM SDK? -------------------------- [](#why-laravel-local-llm-sdk) FeatureDescription**Zero Cloud Dependency**Run LLMs entirely on your local infrastructure**Enterprise Ready**Type-safe, tested, and PSR-compliant**Modern PHP 8.4**Strict typing, readonly classes, union types**Production Tested**Comprehensive test suite with 100% coverage goalsQuick Links ----------- [](#quick-links) - [๐Ÿ“˜ Documentation](docs/index.html) - [โš™๏ธ API Reference](#api-reference) - [๐Ÿงช Testing](#testing) - [๐Ÿš€ Roadmap](#roadmap) - [๐Ÿ“ฆ Packagist](https://packagist.org/packages/laravel-local-llm/sdk) Overview -------- [](#overview) Laravel Local LLM SDK is a modern, enterprise-grade Laravel package designed to integrate local Large Language Models (LLMs) such as Ollama and LM Studio into your Laravel applications. Features -------- [](#features) - **Multi-Driver Architecture** - Support for Ollama, LM Studio, AirLLMLlama, and OpenAI-compatible local servers - **Intelligent Failover** - Automatic fallback to healthy drivers - **Auto-Detection** - Automatically detect available local LLM engines - **Streaming Support** - Server-Sent Events (SSE) for real-time responses - **Token-Based Authentication** - Built-in API token system with rate limiting - **Usage Tracking** - Track token usage and quotas - **Builder Pattern** - Fluent API for building requests - **Event-Driven** - Dispatch events for observability - **Embeddings Support** - Generate vector embeddings for semantic search - **Tool Calling** - Define and use tools/functions with LLMs - **Batch Processing** - Process multiple requests efficiently - **Webhooks** - Send LLM events to external services - **Metrics** - Prometheus-compatible metrics for monitoring - **Caching** - Cache model lists and health status Requirements ------------ [](#requirements) - PHP 8.4+ - Laravel 12+ - Composer 2+ - Local LLM engine (Ollama, LM Studio, or OpenAI-compatible server) Installation ------------ [](#installation) ``` composer require laravel-local-llm/sdk ``` Configuration ------------- [](#configuration) Publish the configuration file: ``` php artisan vendor:publish --provider="LaravelLocalLlm\LocalLlmServiceProvider" --tag="llm-config" ``` ### Environment Variables [](#environment-variables) ``` # Default driver LLM_DEFAULT_DRIVER=ollama # Ollama LLM_OLLAMA_ENABLED=true LLM_OLLAMA_URL=http://localhost:11434 LLM_OLLAMA_DEFAULT_MODEL=llama3.2 # LM Studio LLM_LMSTUDIO_ENABLED=true LLM_LMSTUDIO_URL=http://localhost:1234/v1 LLM_LMSTUDIO_DEFAULT_MODEL=llama-3.2-1b-instruct # OpenAI Compatible LLM_OPENAI_COMPATIBLE_ENABLED=false LLM_OPENAI_COMPATIBLE_URL=http://localhost:8080/v1 # Failover LLM_FAILOVER_ENABLED=true # Auto-detection LLM_AUTO_DETECT=true ``` Usage ----- [](#usage) ### Using the Facade [](#using-the-facade) ``` use LaravelLocalLlm\Facades\LocalLlm; use LaravelLocalLlm\DTO\Message; // Simple chat $response = LocalLlm::chat( new ChatRequest( model: 'llama3.2', messages: [ Message::user('Hello, how are you?'), ] ) ); echo $response->content; ``` ### Using the Builder [](#using-the-builder) ``` $response = LocalLlm::chatWithBuilder() ->model('llama3.2') ->withUserMessage('Hello, how are you?') ->temperature(0.7) ->send(); echo $response->content; ``` ### Streaming [](#streaming) ``` LocalLlm::chatWithBuilder() ->model('llama3.2') ->withUserMessage('Tell me a story') ->stream(true) ->sendStream(function ($chunk) { echo $chunk->content; if ($chunk->finished) { echo "\nDone!\n"; } }); ``` ### Using Specific Driver [](#using-specific-driver) ``` $response = LocalLlm::chat( new ChatRequest(...), Driver::LM_STUDIO ); ``` ### Failover [](#failover) ``` $response = LocalLlm::chatWithFailover(new ChatRequest(...)); ``` ### Checking Models [](#checking-models) ``` $models = LocalLlm::models(); ``` ### Health Check [](#health-check) ``` $isHealthy = LocalLlm::health(); ``` Embeddings ---------- [](#embeddings) Generate vector embeddings for semantic search: ``` use LaravelLocalLlm\Facades\LocalLlm; use LaravelLocalLlm\DTO\EmbeddingRequest; // Single text $response = LocalLlm::embeddings(new EmbeddingRequest( model: 'text-embedding-3-small', input: 'Hello world' )); $embedding = $response->embeddings[0]->embedding; // Multiple texts $response = LocalLlm::embeddings(new EmbeddingRequest( model: 'text-embedding-3-small', input: ['Hello world', 'Goodbye world'] )); ``` Batch Processing ---------------- [](#batch-processing) Process multiple chat requests efficiently: ``` use LaravelLocalLlm\Facades\LocalLlm; use LaravelLocalLlm\DTO\BatchChatRequest; use LaravelLocalLlm\DTO\ChatRequest; use LaravelLocalLlm\DTO\Message; $requests = [ new ChatRequest(model: 'llama3.2', messages: [Message::user('Hello')]), new ChatRequest(model: 'llama3.2', messages: [Message::user('How are you?')]), new ChatRequest(model: 'llama3.2', messages: [Message::user('Tell me a joke')]), ]; $batchResponse = LocalLlm::batchChat(new BatchChatRequest($requests)); echo "Total requests: " . $batchResponse->count(); echo "Total tokens: " . $batchResponse->totalTokens(); echo "Avg latency: " . $batchResponse->averageLatencyMs() . "ms"; ``` Token Authentication -------------------- [](#token-authentication) ### Creating Tokens [](#creating-tokens) ``` use LaravelLocalLlm\Models\LlmToken; use Illuminate\Support\Facades\Hash; $token = LlmToken::create([ 'name' => 'API Token', 'hashed_token' => Hash::make('your-secret-token'), 'abilities' => ['chat', 'stream'], 'rate_limit' => 60, 'monthly_quota' => 1000000, ]); ``` ### Using Tokens [](#using-tokens) Include the token in your request: ``` curl -H "Authorization: Bearer your-secret-token" \ https://your-app.com/api/llm/chat ``` ### Middleware [](#middleware) Protect your routes: ``` Route::middleware(['llm.guard:chat,stream'])->group(function () { Route::post('/llm/chat', [LlmController::class, 'chat']); }); ``` Events ------ [](#events) - `ChatCompleted` - Dispatched when a chat request completes - `StreamChunkReceived` - Dispatched for each streaming chunk ``` Event::listen(\LaravelLocalLlm\Events\ChatCompleted::class, function ($event) { log::info('Chat completed', [ 'model' => $event->response->model, 'latency' => $event->response->latencyMs, ]); }); ``` Webhooks -------- [](#webhooks) Send LLM events to external services: ``` use LaravelLocalLlm\Webhooks\WebhookDispatcher; $webhook = new WebhookDispatcher(); $webhook->register('chat.completed', 'https://your-app.com/webhooks/llm', [ 'secret' => env('WEBHOOK_SECRET'), ]); $webhook->dispatchChatCompleted($request, $response, $driver); ``` Metrics ------- [](#metrics) Track LLM usage with Prometheus-compatible metrics: ``` use LaravelLocalLlm\Services\Metrics; $metrics = new Metrics(); $metrics->recordRequest('ollama', 'llama3.2', 150.5, 20, 50); $metrics->recordRequest('ollama', 'llama3.2', 120.0, 15, 45); // Get per-model metrics $allMetrics = $metrics->getMetrics(); // Get aggregate metrics $aggregate = $metrics->getAggregateMetrics(); // ['total_requests' => 2, 'avg_latency_ms' => 135.25, ...] // Export to Prometheus format $prometheus = $metrics->toPrometheusFormat(); ``` Helpers ------- [](#helpers) Utility functions for common tasks: ``` use LaravelLocalLlm\Helpers\TokenCalculator; use LaravelLocalLlm\Helpers\ResponseFormatter; // Estimate tokens $tokens = TokenCalculator::estimateTokens('Hello world'); // Calculate cost $cost = TokenCalculator::calculateCost(100, 50, 0.001, 0.002); // Format response $html = ResponseFormatter::markdown('**bold** and *italic*'); // Extract code blocks $codeBlocks = ResponseFormatter::extractCode($markdown); ``` Console Commands ---------------- [](#console-commands) ``` # Check driver health php artisan llm:health # Check specific driver php artisan llm:health --driver=ollama # List models php artisan llm:models # List models for specific driver php artisan llm:models --driver=lmstudio # Clear cache php artisan llm:clear-cache ``` Extending --------- [](#extending) ### Custom Driver [](#custom-driver) ``` use LaravelLocalLlm\Contracts\DriverInterface; use LaravelLocalLlm\Enums\Driver; use LaravelLocalLlm\DTO\ChatRequest; use LaravelLocalLlm\DTO\ChatResponse; class CustomDriver implements DriverInterface { public function getDriver(): Driver { return Driver::OLLAMA; // or new Driver('custom') } public function chat(ChatRequest $request): ChatResponse { // Implementation } public function stream(ChatRequest $request, callable $onChunk): void { // Implementation } public function models(): array { // Implementation } public function health(): bool { // Implementation } public function isEnabled(): bool { return true; } } ``` Testing ------- [](#testing) ``` composer test ``` License ------- [](#license) MIT License - see [LICENSE](LICENSE) for details. ### Health Score 37 โ€” LowBetter than 83% of packages Maintenance87 Actively maintained with recent releases Popularity0 Limited adoption so far Community6 Small or concentrated contributor base Maturity47 Maturing project, gaining track record Bus Factor1 Top contributor holds 100% of commits โ€” single point of failure How is this calculated?**Maintenance (25%)** โ€” Last commit recency, latest release date, and issue-to-star ratio. Uses a 2-year decay window. **Popularity (30%)** โ€” Total and monthly downloads, GitHub stars, and forks. Logarithmic scaling prevents top-heavy scores. **Community (15%)** โ€” Contributors, dependents, forks, watchers, and maintainers. Measures real ecosystem engagement. **Maturity (30%)** โ€” Project age, version count, PHP version support, and release stability. ### Release Activity Cadence Every ~0 days Total 2 Last Release 64d ago ### Community Maintainers ![](https://www.gravatar.com/avatar/40be284b8dd88b2d48b0d446df171ae05f8eaaf44ab18b948766c6fafc76f9f4?d=identicon)[shamimstack](/maintainers/shamimstack) --- Top Contributors [![shamimlaravel](https://avatars.githubusercontent.com/u/266431424?v=4)](https://github.com/shamimlaravel "shamimlaravel (5 commits)") --- Tags laravelaiopenaillmollamalm-studiolocal-ai ### Code Quality TestsPHPUnit Static AnalysisPHPStan Type Coverage Yes ### Embed Badge ![Health badge](/badges/laravel-local-llm-sdk/health.svg) ``` [![Health](https://phpackages.com/badges/laravel-local-llm-sdk/health.svg)](https://phpackages.com/packages/laravel-local-llm-sdk) ``` ### Alternatives [cognesy/instructor-php The complete AI toolkit for PHP: unified LLM API, structured outputs, agents, and coding agent control 310107.9k1](/packages/cognesy-instructor-php) PHPackages ยฉ 2026 [Directory](/)[Categories](/categories)[Trending](/trending)[Changelog](/changelog)[Analyze](/analyze)