Promptorum in Cache Repositio Praebitoris Conscia

Superdav AI Agent v1.12.0 introducit promptorum in cache repositio praebitoris conscia, quae sumptus API et moram optimat prompta apud diversos LLM praebitores in cache reponendo. Unicuique praebitori diversa sunt machinae et configurationes cache.

Conspectus

Promptorum in cache repositio tibi sinit:

• Magna prompta, saepe adhibita, in cache reponere
• Sumptus API minuere processu redundanti vitato
• Moram emendare petitionibus in cache conditis
• Vitam cache explicite administrare

Diversi praebitores caching diversimode efficiunt:

• Google Gemini: cachedContents API
• Azure OpenAI: Promptorum caching cum TTL
• OpenRouter: Caching proprius praebitori
• Vertex Anthropic: Promptorum caching cum cache control

Google Gemini: cachedContents API

Google Gemini administrationem cache explicitam per cachedContents API praebet.

Configuratio

$config = [
    'provider' => 'google-gemini',
    'model' => 'gemini-2.0-flash',
    'caching' => [
        'enabled' => true,
        'ttl' => 3600, // 1 hour in seconds
        'max_tokens' => 1000000, // Max tokens to cache
    ],
];

Promptum in Cache Conditum Creare

use Superdav\AI\Providers\GoogleGemini;

$gemini = new GoogleGemini( $config );

$cached_content = $gemini->create_cached_content(
    [
        'system_prompt' => 'You are a helpful assistant...',
        'context' => 'Large context document...',
        'ttl' => 3600,
    ]
);

// Returns: ['cache_id' => 'abc123', 'expires_at' => timestamp]

Prompto in Cache Condito Uti

$response = $gemini->generate(
    [
        'cache_id' => 'abc123',
        'prompt' => 'User question here',
    ]
);

Vita Cache

// List cached contents
$caches = $gemini->list_cached_contents();

// Get cache details
$cache = $gemini->get_cached_content( 'abc123' );

// Extend cache TTL
$gemini->update_cached_content(
    'abc123',
    ['ttl' => 7200] // Extend to 2 hours
);

// Delete cache
$gemini->delete_cached_content( 'abc123' );

Optimae Rationes pro Gemini

• TTL aptum constitue: Aequilibra compendia sumptuum contra vetustatem cache
• Prompta systematis in cache repone: Idem promptum systematis per petitiones reuse
• Usum cache observa: Indaga quae caches maxime adhibeantur
• Caches exspiratas purga: Caches non adhibitas periodice dele

Azure OpenAI: Promptorum Caching

Azure OpenAI promptorum caching cum administratione TTL automatica sustinet.

Configuratio

$config = [
    'provider' => 'azure-openai',
    'model' => 'gpt-4-turbo',
    'api_version' => '2024-08-01-preview',
    'caching' => [
        'enabled' => true,
        'cache_control' => 'max_age=3600',
    ],
];

Caching Activare

use Superdav\AI\Providers\AzureOpenAI;

$azure = new AzureOpenAI( $config );

$response = $azure->generate(
    [
        'system_prompt' => 'You are a helpful assistant...',
        'context' => 'Large context document...',
        'prompt' => 'User question here',
        'cache_control' => 'max_age=3600',
    ]
);

// Response includes cache usage:
// [
//     'content' => '...',
//     'cache_creation_input_tokens' => 1000,
//     'cache_read_input_tokens' => 500,
// ]

Capita Cache

Azure OpenAI capitibus HTTP ad cache control utitur:

Cache-Control: max_age=3600

Valores sustentati:

• max_age=<seconds>: In cache repone per durationem definitam
• no_cache: Hanc petitionem in cache ne repone
• no_store: Neque in cache repone neque reuse

Usum Cache Observare

$response = $azure->generate( [...] );

$cache_tokens = $response['cache_creation_input_tokens'] ?? 0;
$cache_hits = $response['cache_read_input_tokens'] ?? 0;

echo "Cache creation: $cache_tokens tokens\n";
echo "Cache hits: $cache_hits tokens\n";

Optimae Rationes pro Azure OpenAI

• Promptis constantibus utere: Prompta identica ex caching prosunt
• TTL rationabile constitue: Aequilibra sumptum contra recentiam
• Mensuras cache observa: Creationem cache contra hits indaga
• Petitiones similes congloba: Petitiones congrega ut cache hits maxime augeantur

OpenRouter: Caching Proprium Praebitori

OpenRouter caching per praebitores subiectos (OpenAI, Anthropic, etc.) sustinet.

Configuratio

$config = [
    'provider' => 'openrouter',
    'model' => 'openai/gpt-4-turbo',
    'caching' => [
        'enabled' => true,
        'provider_cache' => 'openai', // Use OpenAI's caching
    ],
];

OpenRouter Caching Uti

use Superdav\AI\Providers\OpenRouter;

$router = new OpenRouter( $config );

$response = $router->generate(
    [
        'system_prompt' => 'You are a helpful assistant...',
        'context' => 'Large context document...',
        'prompt' => 'User question here',
        'cache_control' => 'max_age=3600',
    ]
);

Optiones Propriae Praebitori

Diversis praebitoribus diversae sunt machinae caching:

// OpenAI-compatible caching
$response = $router->generate(
    [
        'model' => 'openai/gpt-4-turbo',
        'cache_control' => 'max_age=3600',
    ]
);

// Anthropic-compatible caching
$response = $router->generate(
    [
        'model' => 'anthropic/claude-3-opus',
        'cache_control' => [
            'type' => 'ephemeral',
            'max_tokens' => 1000000,
        ],
    ]
);

Optimae Rationes pro OpenRouter

• Caching praebitoris tui cognosce: Unicuique praebitori diversae sunt machinae
• Mores caching experire: Confirma caching cum praebitore electo tuo operari
• Sumptus observa: Compendia ex caching indaga
• Modelis constantibus utere: Mutatio modelorum cache hits frangit

Vertex Anthropic: Promptorum Caching cum Cache Control

Vertex Anthropic (Google Cloud) promptorum caching cum cache control explicito sustinet.

Configuratio

$config = [
    'provider' => 'vertex-anthropic',
    'model' => 'claude-3-opus',
    'project_id' => 'your-gcp-project',
    'region' => 'us-central1',
    'caching' => [
        'enabled' => true,
        'cache_control' => [
            'type' => 'ephemeral',
            'max_tokens' => 1000000,
        ],
    ],
];

Usus Vertex Anthropic Caching

use Superdav\AI\Providers\VertexAnthropic;

$vertex = new VertexAnthropic( $config );

$response = $vertex->generate(
    [
        'system_prompt' => 'You are a helpful assistant...',
        'context' => 'Large context document...',
        'prompt' => 'User question here',
        'cache_control' => [
            'type' => 'ephemeral',
            'max_tokens' => 1000000,
        ],
    ]
);

// Response includes cache metrics:
// [
//     'content' => '...',
//     'usage' => [
//         'input_tokens' => 1000,
//         'cache_creation_input_tokens' => 500,
//         'cache_read_input_tokens' => 300,
//     ],
// ]

Genera Gubernationis Cache

• ephemeral: In cache serva per durationem petitionis (praedefinitum)
• persistent: In cache serva per plures petitiones (si sustentatur)

Observatio Usus Cache

$response = $vertex->generate( [...] );

$usage = $response['usage'];
$cache_created = $usage['cache_creation_input_tokens'] ?? 0;
$cache_read = $usage['cache_read_input_tokens'] ?? 0;

echo "Cache created: $cache_created tokens\n";
echo "Cache read: $cache_read tokens\n";

Optimae Rationes pro Vertex Anthropic

• Utere caching ephemeral: Bonum ad caching unius sessionis
• Constitue max_tokens apte: Aequilibra magnitudinem cache contra sumptum
• Observa metrica cache: Persequere efficaciam cache
• Experire cum tuo onere laboris: Confirma caching tuo casui usus prodesse

Consilium Caching Inter Provisores

Configuratio Unificata

$config = [
    'caching' => [
        'enabled' => true,
        'default_ttl' => 3600,
        'providers' => [
            'google-gemini' => [
                'ttl' => 3600,
                'max_tokens' => 1000000,
            ],
            'azure-openai' => [
                'cache_control' => 'max_age=3600',
            ],
            'vertex-anthropic' => [
                'cache_control' => [
                    'type' => 'ephemeral',
                    'max_tokens' => 1000000,
                ],
            ],
        ],
    ],
];

Deprehensio Provisoris

$provider = $config['provider'];

$cache_config = $config['caching']['providers'][ $provider ]
    ?? $config['caching'];

// Use provider-specific caching configuration

Consilium Subsidarium

try {
    // Try caching with primary provider
    $response = $primary_provider->generate( $request );
} catch ( CacheException $e ) {
    // Fall back to non-cached request
    $response = $primary_provider->generate(
        array_merge( $request, ['cache_control' => 'no_cache'] )
    );
}

Optimizatio Sumptus

Computa Compendia

$cache_created_tokens = $response['cache_creation_input_tokens'] ?? 0;
$cache_read_tokens = $response['cache_read_input_tokens'] ?? 0;
$regular_tokens = $response['input_tokens'] ?? 0;

// Typical pricing (varies by provider):
$cache_creation_cost = $cache_created_tokens * 0.00001; // 10x cheaper
$cache_read_cost = $cache_read_tokens * 0.000001; // 100x cheaper
$regular_cost = $regular_tokens * 0.00001;

$total_cost = $cache_creation_cost + $cache_read_cost + $regular_cost;
$savings = ($regular_tokens * 0.00001) - $total_cost;

echo "Estimated savings: \$$savings\n";

Monita Optimizationis

• In cache serva magnos system prompts: Maxima compendia sumptus
• Iterum utere contextu: In cache serva documenta contextus saepe adhibita
• Congrega petitiones: Coniunge petitiones similes ut ictus cache maximes
• Observa efficaciam cache: Persequere compendia vera
• Accommoda TTL: Aequilibra sumptum contra recentiam

Solutio Difficultatum

Cache non adhibetur

• Confirma caching in configuratione activatum esse
• Reprehende prompts identica esse (caching exactam congruentiam requirit)
• Confirma cache nondum exspiravisse
• Reprehende limites cache proprios provisoris

Creatio cache deficit

• Confirma magnitudinem cache intra limites provisoris esse
• Reprehende syntaxin gubernationis cache rectam esse
• Cura ut provisor caching pro tuo exemplari sustineat
• Recense documenta provisoris de limitationibus

Sumptus inopinati

• Observa creationem cache contra tokens lectionis cache
• Confirma cache re vera adhiberi
• Reprehende lapsus cache propter variationes prompt
• Considera TTL aut consilium cache accommodare

Comparatio Provisorum

Proprietas	Gemini	Azure OpenAI	OpenRouter	Vertex Anthropic
API Cache	cachedContents	HTTP headers	Provisori proprium	Gubernatio cache
Gubernatio TTL	Expressa	Per headers	Ex provisore pendens	Ephemeral/persistent
Maxima magnitudo cache	1M tokens	Ex provisore pendens	Ex provisore pendens	1M tokens
Imminutio sumptus	90%	90%	Ex provisore pendens	90%
Observatio	Singillatim	Per metrica	Ex provisore pendens	Per usum

Gradus Proximi

1. Elige tuum provisorem: Selige secundum tua necessaria
2. Configura caching: Constitue caching provisori proprium
3. Experire caching: Confirma id cum tuis prompts operari
4. Observa usum: Persequere ictus cache et compendia sumptus
5. Optimiza: Accommoda TTL et consilium cache secundum eventus