本文為英文版的機器翻譯版本，如內容有任何歧義或不一致之處，概以英文版為準。 # 使用適用於 PHP 的 SDK 的 Amazon Bedrock 執行時期範例 下列程式碼範例示範如何使用 適用於 PHP 的 AWS SDK 搭配 Amazon Bedrock 執行期來執行動作和實作常見案例。 *案例*是向您展示如何呼叫服務中的多個函數或與其他 AWS 服務組合來完成特定任務的程式碼範例。 每個範例均包含完整原始碼的連結，您可在連結中找到如何設定和執行內容中程式碼的相關指示。 **Topics** + [案例](#scenarios) + [Amazon Nova](#amazon_nova) + [Amazon Titan 圖像生成器](#amazon_titan_image_generator) + [Anthropic Claude](#anthropic_claude) + [Stable Diffusion](#stable_diffusion) ## 案例 ### 在 Amazon Bedrock 上調用多個基礎模型 下列程式碼範例示範如何在 Amazon Bedrock 上準備並傳送提示至各種大型語言模型 (LLM) **適用於 PHP 的 SDK** GitHub 上提供更多範例。尋找完整範例，並了解如何在 [AWS 程式碼範例儲存庫](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/php/example_code/bedrock-runtime/#code-examples)中設定和執行。 在 Amazon Bedrock 上調用多個 LLM。 ``` namespace BedrockRuntime; class GettingStartedWithBedrockRuntime { protected BedrockRuntimeService $bedrockRuntimeService; public function runExample() { echo "\n"; echo "---------------------------------------------------------------------\n"; echo "Welcome to the Amazon Bedrock Runtime getting started demo using PHP!\n"; echo "---------------------------------------------------------------------\n"; $bedrockRuntimeService = new BedrockRuntimeService(); $prompt = 'In one paragraph, who are you?'; echo "\nPrompt: " . $prompt; echo "\n\nAnthropic Claude:\n"; echo $bedrockRuntimeService->invokeClaude($prompt); echo "\n---------------------------------------------------------------------\n"; $image_prompt = 'stylized picture of a cute old steampunk robot'; echo "\nImage prompt: " . $image_prompt; echo "\n\nStability.ai Stable Diffusion XL:\n"; $diffusionSeed = rand(0, 4294967295); $style_preset = 'photographic'; $base64 = $bedrockRuntimeService->invokeStableDiffusion($image_prompt, $diffusionSeed, $style_preset); $image_path = $this->saveImage($base64, 'stability.stable-diffusion-xl'); echo "The generated image has been saved to $image_path"; echo "\n\nAmazon Titan Image Generation:\n"; $titanSeed = rand(0, 2147483647); $base64 = $bedrockRuntimeService->invokeTitanImage($image_prompt, $titanSeed); $image_path = $this->saveImage($base64, 'amazon.titan-image-generator-v2'); echo "The generated image has been saved to $image_path"; } private function saveImage($base64_image_data, $model_id): string { $output_dir = "output"; if (!file_exists($output_dir)) { mkdir($output_dir); } $i = 1; while (file_exists("$output_dir/$model_id" . '_' . "$i.png")) { $i++; } $image_data = base64_decode($base64_image_data); $file_path = "$output_dir/$model_id" . '_' . "$i.png"; $file = fopen($file_path, 'wb'); fwrite($file, $image_data); fclose($file); return $file_path; } } ``` + 如需 API 詳細資訊，請參閱《*適用於 PHP 的 AWS SDK API 參考*》中的下列主題。 + [InvokeModel](https://docs.aws.amazon.com/goto/SdkForPHPV3/bedrock-runtime-2023-09-30/InvokeModel) + [InvokeModelWithResponseStream](https://docs.aws.amazon.com/goto/SdkForPHPV3/bedrock-runtime-2023-09-30/InvokeModelWithResponseStream) ## Amazon Nova ### Converse 下列程式碼範例示範如何使用 Bedrock 的 Converse API，將文字訊息傳送至 Amazon Nova。 **適用於 PHP 的 SDK** GitHub 上提供更多範例。尋找完整範例，並了解如何在 [AWS 程式碼範例儲存庫](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/php/example_code/bedrock-runtime#code-examples)中設定和執行。 使用 Bedrock 的 Converse API，將文字訊息傳送至 Amazon Nova。 ``` // Use the Conversation API to send a text message to Amazon Nova. use Aws\BedrockRuntime\BedrockRuntimeClient; use Aws\Exception\AwsException; use RuntimeException; class Converse { public function converse(): string { // Create a Bedrock Runtime client in the AWS Region you want to use. $client = new BedrockRuntimeClient([ 'region' => 'us-east-1', 'profile' => 'default' ]); // Set the model ID, e.g., Amazon Nova Lite. $modelId = 'amazon.nova-lite-v1:0'; // Start a conversation with the user message. $userMessage = "Describe the purpose of a 'hello world' program in one line."; $conversation = [ [ "role" => "user", "content" => [["text" => $userMessage]] ] ]; try { // Send the message to the model, using a basic inference configuration. $response = $client->converse([ 'modelId' => $modelId, 'messages' => $conversation, 'inferenceConfig' => [ 'maxTokens' => 512, 'temperature' => 0.5 ] ]); // Extract and return the response text. $responseText = $response['output']['message']['content'][0]['text']; return $responseText; } catch (AwsException $e) { echo "ERROR: Can't invoke {$modelId}. Reason: {$e->getAwsErrorMessage()}"; throw new RuntimeException("Failed to invoke model: " . $e->getAwsErrorMessage(), 0, $e); } } } $demo = new Converse(); echo $demo->converse(); ``` + 如需 API 詳細資訊，請參閱《適用於 PHP 的 AWS SDK API 參考》**中的 [Converse](https://docs.aws.amazon.com/goto/SdkForPHPV3/bedrock-runtime-2023-09-30/Converse)。 ## Amazon Titan 圖像生成器 ### InvokeModel 下列程式碼範例示範如何在 Amazon Bedrock 上調用 Amazon Titan Image 來產生映像。 **適用於 PHP 的 SDK** GitHub 上提供更多範例。尋找完整範例，並了解如何在 [AWS 程式碼範例儲存庫](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/php/example_code/bedrock-runtime#code-examples)中設定和執行。 使用 Amazon Titan 圖像生成器建立影像。 ``` public function invokeTitanImage(string $prompt, int $seed) { // The different model providers have individual request and response formats. // For the format, ranges, and default values for Titan Image models refer to: // https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-titan-image.html $base64_image_data = ""; try { $modelId = 'amazon.titan-image-generator-v2:0'; $request = json_encode([ 'taskType' => 'TEXT_IMAGE', 'textToImageParams' => [ 'text' => $prompt ], 'imageGenerationConfig' => [ 'numberOfImages' => 1, 'quality' => 'standard', 'cfgScale' => 8.0, 'height' => 512, 'width' => 512, 'seed' => $seed ] ]); $result = $this->bedrockRuntimeClient->invokeModel([ 'contentType' => 'application/json', 'body' => $request, 'modelId' => $modelId, ]); $response_body = json_decode($result['body']); $base64_image_data = $response_body->images[0]; } catch (Exception $e) { echo "Error: ({$e->getCode()}) - {$e->getMessage()}\n"; } return $base64_image_data; } ``` + 如需 API 詳細資訊，請參閱《適用於 PHP 的 AWS SDK API 參考》**中的 [InvokeModel](https://docs.aws.amazon.com/goto/SdkForPHPV3/bedrock-runtime-2023-09-30/InvokeModel)。 ## Anthropic Claude ### InvokeModel 下列程式碼範例示範如何使用調用模型 API，將文字訊息傳送至 Anthropic Claude。 **適用於 PHP 的 SDK** GitHub 上提供更多範例。尋找完整範例，並了解如何在 [AWS 程式碼範例儲存庫](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/php/example_code/bedrock-runtime#code-examples)中設定和執行。 調用 Anthropic Claude 2 基礎模型以產生文字。 ``` public function invokeClaude($prompt) { // The different model providers have individual request and response formats. // For the format, ranges, and default values for Anthropic Claude, refer to: // https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-claude.html $completion = ""; try { $modelId = 'anthropic.claude-3-haiku-20240307-v1:0'; // Claude requires you to enclose the prompt as follows: $body = [ 'anthropic_version' => 'bedrock-2023-05-31', 'max_tokens' => 512, 'temperature' => 0.5, 'messages' => [[ 'role' => 'user', 'content' => $prompt ]] ]; $result = $this->bedrockRuntimeClient->invokeModel([ 'contentType' => 'application/json', 'body' => json_encode($body), 'modelId' => $modelId, ]); $response_body = json_decode($result['body']); $completion = $response_body->content[0]->text; } catch (Exception $e) { echo "Error: ({$e->getCode()}) - {$e->getMessage()}\n"; } return $completion; } ``` + 如需 API 詳細資訊，請參閱《適用於 PHP 的 AWS SDK API 參考》**中的 [InvokeModel](https://docs.aws.amazon.com/goto/SdkForPHPV3/bedrock-runtime-2023-09-30/InvokeModel)。 ## Stable Diffusion ### InvokeModel 下列程式碼範例示範如何在 Amazon Bedrock 上調用 Stability.ai Stable Diffusion XL 以產生映像。 **適用於 PHP 的 SDK** GitHub 上提供更多範例。尋找完整範例，並了解如何在 [AWS 程式碼範例儲存庫](https://github.com/awsdocs/aws-doc-sdk-examples/tree/main/php/example_code/bedrock-runtime#code-examples)中設定和執行。 使用 Stable Diffusion 建立映像。 ``` public function invokeStableDiffusion(string $prompt, int $seed, string $style_preset) { // The different model providers have individual request and response formats. // For the format, ranges, and available style_presets of Stable Diffusion models refer to: // https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-stability-diffusion.html $base64_image_data = ""; try { $modelId = 'stability.stable-diffusion-xl-v1'; $body = [ 'text_prompts' => [ ['text' => $prompt] ], 'seed' => $seed, 'cfg_scale' => 10, 'steps' => 30 ]; if ($style_preset) { $body['style_preset'] = $style_preset; } $result = $this->bedrockRuntimeClient->invokeModel([ 'contentType' => 'application/json', 'body' => json_encode($body), 'modelId' => $modelId, ]); $response_body = json_decode($result['body']); $base64_image_data = $response_body->artifacts[0]->base64; } catch (Exception $e) { echo "Error: ({$e->getCode()}) - {$e->getMessage()}\n"; } return $base64_image_data; } ``` + 如需 API 詳細資訊，請參閱《適用於 PHP 的 AWS SDK API 參考》**中的 [InvokeModel](https://docs.aws.amazon.com/goto/SdkForPHPV3/bedrock-runtime-2023-09-30/InvokeModel)。