How to manage large data sets efficiently using Laravel’s chunk() and cursor() methods?

Laravel's chunk() processes records in fixed-size batches (e.g., 1000 at a time) using LIMIT/OFFSET, while cursor() streams one record at a time via generators for minimal memory use on massive datasets.​

Use chunk(1000, callback) for batch operations like exports or jobs where you need related data with with(); switch to cursor() for memory-critical tasks like 1M+ record updates since it avoids loading collections entirely. Always pair with select() for essential columns only and chunkById() on mutable tables to prevent skips/duplicates.

Code

// chunk() - batch processing (medium datasets)
User::select('id', 'email')->chunk(1000, function ($users) {
    foreach ($users as $user) {
        $user->update(['last_login' => now()]);
    }
});

// cursor() - memory efficient streaming (large datasets)
foreach (User::select('id', 'email')->cursor() as $user) {
    // Process one by one - no memory buildup
    processUser($user);
}

// chunkById() - safe for changing tables
User::chunkById(500, function ($users) {
    // Stable ID-based chunks
});