Prompt Caching nke Na-amata Provider

Superdav AI Agent v1.12.0 na-ewebata prompt caching nke na-amata provider, nke na-eme ka ụgwọ API na latency ka mma site n’ịchekwa prompts gafee providers LLM dị iche iche. Provider ọ bụla nwere usoro caching na nhazi dị iche iche.

Nchịkọta

Prompt caching na-enye gị ohere:

Chekwaa prompts buru ibu, a na-ejikarị eme ihe
Belata ụgwọ API site n’izere nhazi ugboro ugboro na-enweghị mkpa
Mee ka latency ka mma maka arịrịọ echekwara na cache
Jikwaa cache lifecycle n'ụzọ doro anya

Providers dị iche iche na-etinye caching n’ọrụ n’ụzọ dị iche:

Google Gemini: cachedContents API
Azure OpenAI: Prompt caching nwere TTL
OpenRouter: Caching pụrụ iche maka provider
Vertex Anthropic: Prompt caching nwere cache control

Google Gemini: cachedContents API

Google Gemini na-enye njikwa cache doro anya site na cachedContents API.

Nhazi

$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
    ],
];

Ịmepụta Prompt Echekwara na Cache

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]

Iji Prompt Echekwara na Cache

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

Cache Lifecycle

// 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' );

Omume Kachasị Mma maka Gemini

Tọọ TTL kwesịrị ekwesị: Hazie n’etiti ịchekwa ụgwọ na ịka nká nke cache
Chekwaa system prompts na cache: Jiri otu system prompt ọzọ gafee arịrịọ
Nyochaa ojiji cache: Soro caches a na-eji nke ukwuu
Hichapụ caches gwụchara oge: Hichapụ caches anaghị eji eme ihe kwa oge

Azure OpenAI: Prompt Caching

Azure OpenAI na-akwado prompt caching nwere njikwa TTL akpaka.

Nhazi

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

Ime ka Caching rụọ ọrụ

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,
// ]

Cache Headers

Azure OpenAI na-eji HTTP headers maka cache control:

Cache-Control: max_age=3600

Ụkpụrụ akwadoro:

max_age=<seconds>: Chekwaa na cache ruo oge akọwapụtara
no_cache: Echekwala arịrịọ a na cache
no_store: Echekwala na cache ma ejikwala ọzọ

Nyochaa Ojiji Cache

$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";

Omume Kachasị Mma maka Azure OpenAI

Jiri prompts na-agbanwebeghị: Prompts yiri otu na-erite uru na caching
Tọọ TTL ezi uche dị na ya: Hazie n’etiti ụgwọ na ịdị ọhụrụ
Nyochaa metrics cache: Soro imepụta cache megide hits
Kpokọta arịrịọ yiri ibe ha: Kpọkọta arịrịọ iji mee ka cache hits baa ụba

OpenRouter: Caching Pụrụ Iche maka Provider

OpenRouter na-akwado caching site na providers dị n’okpuru (OpenAI, Anthropic, wdg.).

Nhazi

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

Iji OpenRouter Caching

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',
    ]
);

Nhọrọ Pụrụ Iche maka Provider

Providers dị iche iche nwere usoro caching dị iche:

// 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,
        ],
    ]
);

Omume Kachasị Mma maka OpenRouter

Mara caching nke provider gị: Provider ọ bụla nwere usoro dị iche
Nwale omume caching: Nyochaa na caching na-arụ ọrụ na provider ị họọrọ
Nyochaa ụgwọ: Soro ego echekwara site na caching
Jiri models na-agbanwebeghị: Ịgbanwe models na-emebi cache hits

Vertex Anthropic: Prompt Caching nwere Cache Control

Vertex Anthropic (Google Cloud) na-akwado prompt caching nwere cache control doro anya.

Nhazi

$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,
        ],
    ],
];

Iji 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,
//     ],
// ]

Ụdị Cache Control

ephemeral: Cache maka oge arịrịọ ahụ na-adị (ndabara)
persistent: Cache gafee ọtụtụ arịrịọ (ma ọ bụrụ na a na-akwado ya)

Nlekọta Ojiji 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";

Omume Kachasị Mma maka Vertex Anthropic

Jiri ephemeral caching: Ọ dị mma maka caching nke otu nnọkọ
Tọọ max_tokens nke ọma: Mee nha cache na ọnụ ahịa kwekọọ
Nyochaa cache metrics: Soro ịdị irè cache
Nwale na ibu ọrụ gị: Kwenye na caching bara uru maka ọnọdụ ojiji gị

Atụmatụ Caching Gafee Provider

Nhazi Jikọrọ Ọnụ

$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,
                ],
            ],
        ],
    ],
];

Nchọpụta Provider

$provider = $config['provider'];

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

// Use provider-specific caching configuration

Atụmatụ Fallback

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'] )
    );
}

Ibelata Ọnụ Ahịa

Gbakọọ Nchekwa Ego

$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";

Ndụmọdụ Optimization

Cache nnukwu system prompts: Nchekwa ego kachasị ukwuu
Jigharịa context: Cache akwụkwọ context a na-ejikarị
Mee arịrịọ n’ìgwè: Kpokọta arịrịọ yiri ibe ha iji mee ka cache hits dị elu
Nyochaa ịdị irè cache: Soro ezigbo ego echekwara
Gbanwee TTL: Mee ọnụ ahịa na ịdị ọhụrụ kwekọọ

Ịchọpụta na idozi nsogbu

A naghị eji cache eme ihe

Kwenye na enyere caching aka na nhazi
Lelee na prompts bụ otu kpọmkwem (caching chọrọ nkwekọrịta zuru ezu)
Kwenye na cache agafeghị oge
Lelee oke cache pụrụ iche nke provider

Ịmepụta cache na-ada

Kwenye na nha cache dị n’ime oke provider
Lelee na syntax cache control ziri ezi
Gbaa mbọ hụ na provider na-akwado caching maka model gị
Nyochaa documentation provider maka mmachi

Ọnụ ahịa a na-atụghị anya ya

Nyochaa imepụta cache megide tokens a na-agụ site na cache
Kwenye na a na-eji cache n’ezie
Lelee cache misses n’ihi mgbanwe prompt
Tụlee ịgbanwe TTL ma ọ bụ atụmatụ cache

Ntụnyere Provider

Njirimara	Gemini	Azure OpenAI	OpenRouter	Vertex Anthropic
Cache API	cachedContents	HTTP headers	Nke provider pụrụ iche	Cache control
Njikwa TTL	Doro anya	Site na headers	Dabere na provider	Ephemeral/persistent
Oke nha cache	tokens 1M	Dabere na provider	Dabere na provider	tokens 1M
Mbelata ọnụ ahịa	90%	90%	Dabere na provider	90%
Nlekọta	Zuru ezu	Site na metrics	Dabere na provider	Site na usage

Nzọụkwụ Ndị Ọzọ

Họrọ provider gị: Họrọ dabere na mkpa gị
Hazie caching: Hazie caching pụrụ iche maka provider
Nwale caching: Kwenye na ọ na-arụ ọrụ na prompts gị
Nyochaa ojiji: Soro cache hits na ego echekwara
Mee ka ọ ka mma: Gbanwee TTL na atụmatụ cache dabere na nsonaazụ

Nchịkọta​

Google Gemini: cachedContents API​

Nhazi​

Ịmepụta Prompt Echekwara na Cache​

Iji Prompt Echekwara na Cache​

Cache Lifecycle​

Omume Kachasị Mma maka Gemini​

Azure OpenAI: Prompt Caching​

Nhazi​

Ime ka Caching rụọ ọrụ​

Cache Headers​

Nyochaa Ojiji Cache​

Omume Kachasị Mma maka Azure OpenAI​

OpenRouter: Caching Pụrụ Iche maka Provider​

Nhazi​

Iji OpenRouter Caching​

Nhọrọ Pụrụ Iche maka Provider​

Omume Kachasị Mma maka OpenRouter​

Vertex Anthropic: Prompt Caching nwere Cache Control​

Nhazi​

Iji Vertex Anthropic Caching​

Ụdị Cache Control​

Nlekọta Ojiji Cache​

Omume Kachasị Mma maka Vertex Anthropic​

Atụmatụ Caching Gafee Provider​

Nhazi Jikọrọ Ọnụ​

Nchọpụta Provider​

Atụmatụ Fallback​

Ibelata Ọnụ Ahịa​

Gbakọọ Nchekwa Ego​

Ndụmọdụ Optimization​

Ịchọpụta na idozi nsogbu​

A naghị eji cache eme ihe​

Ịmepụta cache na-ada​

Ọnụ ahịa a na-atụghị anya ya​

Ntụnyere Provider​

Nzọụkwụ Ndị Ọzọ​

Nchịkọta

Google Gemini: cachedContents API

Nhazi

Ịmepụta Prompt Echekwara na Cache

Iji Prompt Echekwara na Cache

Cache Lifecycle

Omume Kachasị Mma maka Gemini

Azure OpenAI: Prompt Caching

Nhazi

Ime ka Caching rụọ ọrụ

Cache Headers

Nyochaa Ojiji Cache

Omume Kachasị Mma maka Azure OpenAI

OpenRouter: Caching Pụrụ Iche maka Provider

Nhazi

Iji OpenRouter Caching

Nhọrọ Pụrụ Iche maka Provider

Omume Kachasị Mma maka OpenRouter

Vertex Anthropic: Prompt Caching nwere Cache Control

Nhazi

Iji Vertex Anthropic Caching

Ụdị Cache Control

Nlekọta Ojiji Cache

Omume Kachasị Mma maka Vertex Anthropic

Atụmatụ Caching Gafee Provider

Nhazi Jikọrọ Ọnụ

Nchọpụta Provider

Atụmatụ Fallback

Ibelata Ọnụ Ahịa

Gbakọọ Nchekwa Ego

Ndụmọdụ Optimization

Ịchọpụta na idozi nsogbu

A naghị eji cache eme ihe

Ịmepụta cache na-ada

Ọnụ ahịa a na-atụghị anya ya

Ntụnyere Provider

Nzọụkwụ Ndị Ọzọ