Kaydinta Prompt-ka ee Ka Warqabta Provider-ka

Superdav AI Agent v1.12.0 wuxuu soo bandhigayaa kaydinta prompt-ka ee ka warqabta provider-ka, taas oo hagaajisa kharashaadka API iyo daahitaanka iyadoo prompts lagu kaydiyo LLM providers kala duwan. Provider kasta wuxuu leeyahay habab kaydin iyo dejinno kala duwan.

Dulmar

Kaydinta prompt-ku waxay kuu oggolaanaysaa inaad:

• Kaydiso prompts waaweyn oo si joogto ah loo isticmaalo
• Yareyso kharashaadka API adigoo ka fogaanaya habayn soo noqnoqota
• Hagaajiso daahitaanka codsiyada kaydsan
• Si cad u maamusho wareegga nolosha kaydka

Providers kala duwan si kala duwan ayay u hirgeliyaan kaydinta:

• Google Gemini: cachedContents API
• Azure OpenAI: Kaydinta prompt-ka oo leh TTL
• OpenRouter: Kaydin u gaar ah provider-ka
• Vertex Anthropic: Kaydinta prompt-ka oo leh xakamaynta kaydka

Google Gemini: cachedContents API

Google Gemini wuxuu bixiyaa maareyn kayd oo cad iyadoo loo marayo cachedContents API.

Dejinta

$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
    ],
];

Abuuridda Prompt Kaydsan

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]

Isticmaalka Prompt Kaydsan

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

Wareegga Nolosha Kaydka

// 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' );

Hab-dhaqannada Ugu Fiican ee Gemini

• Deji TTL ku habboon: Isku dheellitir kaydinta kharashka iyo duugowga kaydka
• Kaydi system prompts: Dib u isticmaal isla system prompt-ka codsiyada oo dhan
• La soco isticmaalka kaydka: Raadraac kaydadka ugu isticmaalka badan
• Nadiifi kaydadka dhacay: Si xilliyeed u tirtir kaydadka aan la isticmaalin

Azure OpenAI: Kaydinta Prompt-ka

Azure OpenAI wuxuu taageeraa kaydinta prompt-ka oo leh maareyn TTL oo otomaatig ah.

Dejinta

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

Daaridda Kaydinta

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,
// ]

Headers-ka Kaydka

Azure OpenAI wuxuu u adeegsadaa HTTP headers xakamaynta kaydka:

Cache-Control: max_age=3600

Qiimayaasha la taageero:

• max_age=<seconds>: Kaydi muddada la cayimay
• no_cache: Ha kaydin codsigan
• no_store: Ha kaydin hana dib u isticmaalin

La Socodka Isticmaalka Kaydka

$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";

Hab-dhaqannada Ugu Fiican ee Azure OpenAI

• Isticmaal prompts isku mid ah: Prompts isku eg waxay ka faa’iidaystaan kaydinta
• Deji TTL macquul ah: Isku dheellitir kharashka iyo cusaybka
• La soco cabbirrada kaydka: Raadraac abuurista kaydka iyo hits-ka
• Isku ururi codsiyada la midka ah: Kooxee codsiyada si loo kordhiyo cache hits

OpenRouter: Kaydin U Gaar ah Provider-ka

OpenRouter wuxuu taageeraa kaydinta iyadoo loo marayo providers hoose (OpenAI, Anthropic, iwm.).

Dejinta

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

Isticmaalka Kaydinta OpenRouter

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',
    ]
);

Ikhtiyaarrada U Gaar ah Provider-ka

Providers kala duwan waxay leeyihiin habab kaydin oo kala duwan:

// 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,
        ],
    ]
);

Hab-dhaqannada Ugu Fiican ee OpenRouter

• Ogow kaydinta provider-kaaga: Provider kasta wuxuu leeyahay habab kala duwan
• Tijaabi hab-dhaqanka kaydinta: Xaqiiji in kaydintu la shaqayso provider-ka aad dooratay
• La soco kharashaadka: Raadraac kaydka ka yimaada kaydinta
• Isticmaal models isku mid ah: Bedelidda models waxay jebisaa cache hits

Vertex Anthropic: Kaydinta Prompt-ka oo leh Xakamaynta Kaydka

Vertex Anthropic (Google Cloud) wuxuu taageeraa kaydinta prompt-ka oo leh xakamayn kayd oo cad.

Dejinta

$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,
        ],
    ],
];

Isticmaalka 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,
//     ],
// ]

Noocyada Cache Control

• ephemeral: Cache muddada codsiga (caadiga)
• persistent: Cache codsiyo badan dhexdiis (haddii la taageero)

La-socodka Isticmaalka 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";

Hab-dhaqannada Ugu Fiican ee Vertex Anthropic

• Isticmaal ephemeral caching: Ku fiican caching hal-fadhi ah
• U deji max_tokens si habboon: Isku dheelli tir cabbirka cache iyo kharashka
• La soco cabbirrada cache: Raad-raac waxtarka cache
• Ku tijaabi culayskaaga shaqo: Xaqiiji in caching uu faa'iido u leeyahay xaaladdaada isticmaal

Istaraatiijiyadda Caching ee Bixiyeyaasha Kala Duwan

Habayn Midaysan

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

Ogaanshaha Bixiyaha

$provider = $config['provider'];

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

// Use provider-specific caching configuration

Istaraatiijiyadda 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'] )
    );
}

Hagaajinta Kharashka

Xisaabi Kaydka

$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";

Talooyin Hagaajin

• Cache system prompts waaweyn: Kaydka kharashka ugu weyn
• Dib u isticmaal context: Cache dukumiintiyada context ee si joogto ah loo isticmaalo
• Isku ururi codsiyada: Kooxee codsiyada isku midka ah si loo kordhiyo cache hits
• La soco waxtarka cache: Raad-raac kaydka dhabta ah
• Hagaaji TTL: Isku dheelli tir kharashka iyo cusaybka

Xallinta Dhibaatooyinka

Cache lama isticmaalayo

• Xaqiiji in caching lagu daaray habaynta
• Hubi in prompts-ku isku mid yihiin (caching wuxuu u baahan yahay is-waafajin sax ah)
• Xaqiiji in cache uusan dhicin
• Hubi xaddidaadaha cache ee u gaarka ah bixiyaha

Abuurista cache way fashilantay

• Xaqiiji in cabbirka cache uu ku jiro xaddidaadaha bixiyaha
• Hubi in syntax-ka cache control uu sax yahay
• Hubi in bixiyuhu taageero caching ee model-kaaga
• Dib u eeg dukumiintiyada bixiyaha ee xaddidaadaha

Kharashyo aan la filayn

• La soco abuurista cache marka loo eego cache read tokens
• Xaqiiji in cache dhab ahaantii la isticmaalayo
• Hubi cache misses ay sababaan kala duwanaanshaha prompt
• Tixgeli hagaajinta TTL ama istaraatiijiyadda cache

Isbarbardhigga Bixiyeyaasha

Astaanta	Gemini	Azure OpenAI	OpenRouter	Vertex Anthropic
Cache API	cachedContents	HTTP headers	U gaar ah bixiyaha	Cache control
Xakamaynta TTL	Cad	Iyadoo loo marayo headers	Ku xiran bixiyaha	Ephemeral/persistent
Cabbirka ugu badan ee cache	1M tokens	Ku xiran bixiyaha	Ku xiran bixiyaha	1M tokens
Dhimista kharashka	90%	90%	Ku xiran bixiyaha	90%
La-socodka	Faahfaahsan	Iyadoo loo marayo metrics	Ku xiran bixiyaha	Iyadoo loo marayo usage

Tallaabooyinka Xiga

1. Dooro bixiyahaaga: Ku dooro iyadoo lagu salaynayo baahiyahaaga
2. Habeey caching: Diyaari caching u gaar ah bixiyaha
3. Tijaabi caching: Xaqiiji inuu la shaqeeyo prompts-kaaga
4. La soco isticmaalka: Raad-raac cache hits iyo kaydka kharashka
5. Hagaaji: Hagaaji TTL iyo istaraatiijiyadda cache iyadoo lagu salaynayo natiijooyinka