完整的嵌入请求和响应架构

完整的同步架构

{
    "schemaVersion": "nova-multimodal-embed-v1",
    "taskType": "SINGLE_EMBEDDING",
    "singleEmbeddingParams": {
        "embeddingPurpose": "GENERIC_INDEX" | "GENERIC_RETRIEVAL" | "TEXT_RETRIEVAL" | "IMAGE_RETRIEVAL" | "VIDEO_RETRIEVAL" | "DOCUMENT_RETRIEVAL" | "AUDIO_RETRIEVAL" | "CLASSIFICATION" | "CLUSTERING",
        "embeddingDimension": 256 | 384 | 1024 | 3072,
        "text": {
            "truncationMode": "START" | "END" | "NONE",
            "value": string,
            "source": SourceObject,
        },
        "image": {
            "detailLevel": "STANDARD_IMAGE" | "DOCUMENT_IMAGE",
            "format": "png" | "jpeg" | "gif" | "webp",
            "source": SourceObject
        },
        "audio": {
            "format": "mp3" | "wav" | "ogg",
            "source": SourceObject
        },
        "video": {
            "format": "mp4" | "mov" | "mkv" | "webm" | "flv" | "mpeg" | "mpg" | "wmv" | "3gp",
            "source": SourceObject,
            "embeddingMode": "AUDIO_VIDEO_COMBINED" | "AUDIO_VIDEO_SEPARATE"
        }
    }
}

   

以下列表包含了该请求的所有参数：

InvokeModel 响应正文

当 invokeModel 返回“成功”结果时，响应正文将具有以下结构：

{
   "embeddings": [
      {
          "embeddingType": "TEXT" | "IMAGE" | "VIDEO" | "AUDIO" | "AUDIO_VIDEO_COMBINED",
          "embedding": number[],
          "truncatedCharLength": int // Only included if text input was truncated
      }
    ]                       
}
   

以下列表包含了该响应的所有参数：

完整的异步架构

您可以使用 Amazon Bedrock 运行时 API 函数 StartAsyncInvoke, GetAsyncInvoke 和 ListAsyncInvokes 异步生成嵌入。如果要使用 Nova Embeddings 对长内容（例如长篇文本或长度超过 30 秒的视频和音频）进行分段，则必须使用异步 API。

调用 StartAsyncInvoke 时，必须提供 modelId、outputDataConfig 和 modelInput 参数。

response = bedrock_runtime.start_async_invoke(
    modelId="amazon.nova-2-multimodal-embeddings-v1:0",
    outputDataConfig=Data Config,
    modelInput=Model Input
)
  

outputDataConfig 指定要将生成的输出保存到的 S3 存储桶。它具有以下结构：

{
    "s3OutputDataConfig": {
        "s3Uri": "s3://your-s3-bucket"
    }
} 
  

s3Uri 是目标存储桶的 S3 URI。有关其他可选参数，请参阅 StartAsyncInvoke 文档。

modelInput 参数使用以下结构。

{
    "schemaVersion": "nova-multimodal-embed-v1",
    "taskType": "SEGMENTED_EMBEDDING",
    "segmentedEmbeddingParams": {
        "embeddingPurpose": "GENERIC_INDEX" | "GENERIC_RETRIEVAL" | "TEXT_RETRIEVAL" | "IMAGE_RETRIEVAL" | "VIDEO_RETRIEVAL" | "DOCUMENT_RETRIEVAL" | "AUDIO_RETRIEVAL" | "CLASSIFICATION" | "CLUSTERING",
        "embeddingDimension": 256 | 384 | 1024 | 3072,
        "text": {
            "truncationMode": "START" | "END" | "NONE",
            "value": string,
            "source": {
                "s3Location": {
                    "uri": "s3://Your S3 Object"
                }
            },
            "segmentationConfig": {
                "maxLengthChars": int
            }
        },
        "image": {
            "format": "png" | "jpeg" | "gif" | "webp",
            "source": SourceObject,
            "detailLevel": "STANDARD_IMAGE" | "DOCUMENT_IMAGE"
        },
        "audio": {
            "format": "mp3" | "wav" | "ogg",
            "source": SourceObject,
            "segmentationConfig": {
                "durationSeconds": int
            }
        },
        "video": {
            "format": "mp4" | "mov" | "mkv" | "webm" | "flv" | "mpeg" | "mpg" | "wmv" | "3gp",
            "source": SourceObject,
            "embeddingMode": "AUDIO_VIDEO_COMBINED" | "AUDIO_VIDEO_SEPARATE",
            "segmentationConfig": {
                "durationSeconds": int
            }
        }
    }
}
  

以下列表包含了该请求的所有参数：

StartAsyncInvoke 响应

调用 StartAsyncInvoke 时的响应将具有以下结构。invocationArn 可用于通过 GetAsyncInvoke 函数查询异步作业的状态。

{
    "invocationArn": "arn:aws:bedrock:us-east-1:xxxxxxxxxxxx:async-invoke/lvmxrnjf5mo3",
}
  

异步输出

异步嵌入生成完成后，输出构件将会写入您指定为输出目标的 S3 存储桶。文件的结构如下：

   amzn-s3-demo-bucket/
    job-id/
        segmented-embedding-result.json
        embedding-audio.jsonl
        embedding-image.json
        embedding-text.jsonl
        embedding-video.jsonl
        manifest.json
  

segmented-embedding-result.json 将包含总体作业结果以及对相应 jsonl 文件的引用，后者包含每种模态的实际嵌入。以下是对文件进行截断对示例：

{
    "sourceFileUri": string, 
    "embeddingDimension": 256 | 384 | 1024 | 3072,
    "embeddingResults": [
        {
            "embeddingType": "TEXT" | "IMAGE" | "VIDEO" | "AUDIO" | "AUDIO_VIDEO_COMBINED",
            "status": "SUCCESS" | "FAILURE" | "PARTIAL_SUCCESS",
            "failureReason": string, // Granular error codes
            "message": string, // Human-readbale failure message
            "outputFileUri": string // S3 URI to a "embedding-modality.jsonl" file
        }
        ...
    ]
}
  

embedding-modality.json 将是包含每种模态的嵌入输出的 jsonl 文件。该 jsonl 文件中的每一行都将遵循以下架构：

{
    "embedding": number[], // The generated embedding vector
    "segmentMetadata": {
        "segmentIndex": number,
        "segmentStartCharPosition": number, // Included for text only
        "segmentEndCharPosition": number, // Included for text only
        "truncatedCharLength": number, // Included only when text gets truncated
        "segmentStartSeconds": number, // Included for audio/video only
        "segmentEndSeconds": number // Included for audio/video only
    },
    "status": "SUCCESS" | "FAILURE",
    "failureReason": string, // Granular error codes
    "message": string // Human-readable failure message
}
  

以下列表包含了该响应的所有参数。对于文本字符或音频/视频时间，所有开始和结束时间均从零开始计算。此外，所有结束文本位置或音频/视频时间值都包括本数。

Nova Embeddings 的文件限制

同步操作既可以接受 S3 输入，也可以接受内联块。异步操作只能接受 S3 输入。

异步生成嵌入时，您需要确保将文件分成适当数量的分段。对于文本嵌入，分段不能超过 1900 个。对于音频和视频嵌入，分段不能超过 1434 个。