All Products
Search

This Product

    Intelligent Media Management:Billing Items for Intelligent Media Management

    Document Center

    Intelligent Media Management:Billing Items for Intelligent Media Management

    Last Updated:Mar 10, 2026

    This topic describes the billing items for Intelligent Media Management (IMM) and related details.

    Billing Items for Alibaba Cloud International Website

    Intelligent Media Management (IMM) includes the following billing items: image intelligence, metadata management, media processing, document processing, and file processing.

    Important

    Starting at 11:00 UTC+8 on July 28, 2025, IMM will begin charging for some previously free features and will adjust prices for some existing billing items. For more information, refer to the IMM Pricing Adjustment Notice.

    Image Intelligence

    For detailed pricing of image intelligence billing items, refer to the following table.

    Billing Item

    Billing Item Description

    Related API Operations

    Related x-oss-process Operations

    Price (USD)

    Unit

    ImageDetect

    Face detection

    • DetectImageFaces (face detection)

    • CompareImageFaces (face similarity comparison)

    • SearchImageFigureCluster (query face clusters)

    • CreateFacesSearchingTask (create a similar-face search task)

    • image/faces

    • image/crop,g_face

    • image/blur,g_face

    • image/blur,g_faces

    0.028

    per 1,000 calls

    Human body detection

    DetectImageBodies (human body detection)

    image/bodies

    Vehicle detection

    DetectImageCars

    image/cars

    ImageLabel

    Image tagging

    DetectImageLabels (image label detection)

    image/labels

    0.142

    per 1,000 calls

    ImageFace

    Face images

    CreateFacesSearchingTask (create a similar-face search task)

    0.028

    per 1,000 calls

    ImageFaceClustering

    Face clustering

    • CreateFigureClusteringTask (create a face clustering task)

    • CreateFigureClustersMergingTask (create a face cluster merging task)

    7.0754717

    per 1,000 calls

    GenerateStory

    Story generation

    CreateStory (create a story)

    7.0754717

    per 1,000 calls

    ImageMosaic

    Image mosaic

    AddImageMosaic (add an image mosaic)

    0.0074

    per 1,000 calls

    ImageCropping

    Smart image cropping suggestions

    DetectImageCropping (detect visually optimal cropping regions)

    image/crop,g_auto

    0.1415094

    per 1,000 calls

    ImageQRCodes

    QR code detection in images

    DetectImageCodes (QR code detection)

    image/codes

    0.1132075

    per 1,000 calls

    ImageSplicing

    Image stitching

    CreateImageSplicingTask (create an image stitching task)

    • Output resolution less than 6000 × 4000: 0.745

    • Output resolution greater than or equal to 6000 × 4000: 2.234

    per 1,000 calls

    ImageToPDF

    Image-to-PDF conversion

    CreateImageToPDFTask (create an image-to-PDF conversion task)

    0.0074

    per 1,000 images

    ImageScoring

    Image quality scoring

    DetectImageScore (image quality scoring)

    image/scoring

    0.0424528

    per 1,000 calls

    LocationDateClustering

    Spatiotemporal clustering

    CreateLocationDateClusteringTask (create a spatiotemporal clustering task)

    Free for a limited time

    per 1,000 calls

    SimilarImageClustering

    Image clustering

    CreateSimilarImageClusteringTask (create a similar-image clustering task)

    Free for a limited time

    per 1,000 calls

    Blindwatermark

    Blind watermarking for images

    • EncodeBlindWatermark (add a blind watermark)

    • CreateDecodeBlindWatermarkTask (create a blind watermark decoding task)

    • image/blindwatermark

    • image/deblindwatermark

    0.0990566

    per 1,000 calls

    ReverseGeocoding

    Reverse geocoding

    DetectMediaMeta (retrieve media file metadata)

    Note

    Charged only when the media file contains geographic location information.

    0.1415094

    per 1,000 calls

    ImageTexts

    Optical character recognition (OCR) for images

    DetectImageTexts (OCR for images)

    7.0754717

    per 1,000 calls

    Metadata Management

    For detailed pricing of metadata management billing items, refer to the following table.

    Billing Item

    Billing item description

    Related API Operations

    Related x-oss-process Operations

    Price (USD)

    Unit

    StandardQueryL0

    Basic queries

    • GetFileMeta (retrieve file metadata)

    • DeleteFileMeta (delete file metadata)

    • UpdateFileMeta (update file metadata)

    • BatchDeleteFileMeta (batch delete file metadata)

    • GetFigureCluster (retrieve face cluster information)

    • UpdateFigureCluster (update face clusters)

    • UpdateStory (update a story)

    • GetStory (retrieve story information)

    • DeleteStory (delete a story)

    • UpdateLocationDateCluster (update spatiotemporal clusters)

    • DeleteLocationDateCluster (delete spatiotemporal clusters)

    task/get

    0.001

    per 1,000 calls

    StandardQueryL1

    Standard queries

    • BatchGetFileMeta (batch retrieve file metadata)

    • BatchUpdateFileMeta (batch update file metadata)

    • ListFaceGroups (list face groups in a media set)

    • AddStoryFiles (add files to a story)

    • QueryStories (query stories)

    • RemoveStoryFiles (remove files from a story)

    • QueryFigureClusters (Query Figure Clustering)

    • QueryLocationDateClusters (query spatiotemporal clusters)

    0.002

    per 1,000 calls

    StandardQueryL2

    Advanced queries

    • SimpleQuery (simple query)

    • FuzzyQuery (fuzzy query)

    0.074

    per 1,000 calls

    MediaMeta

    Retrieve media information

    • DetectMediaMeta (retrieve media file metadata)

    • GetMediaMeta (retrieve media file metadata)

    • audio/info

    • video/info

    0.1415094

    per 1,000 calls

    SemanticAnalyze

    Semantic analysis

    SemanticQuery (natural language query)

    0.52

    per 1,000 calls

    Media Processing

    For detailed pricing of media processing billing items, refer to the following table.

    Billing Item

    Billing item description

    Related API Operations

    Related x-oss-process Operations

    Price (USD)

    Unit

    AudioCompress

    Audio transcoding

    CreateMediaConvertTask (create a media transcoding task)

    • audio/concat

    • audio/compress

    0.0000141509

    per second of audio

    VideoCompressCopy

    Container format conversion

    CreateMediaConvertTask (create a media transcoding task)

    0.00001433525

    per second of video

    VideoCompress264LD

    H.264 LD transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0000509434

    Videos per second

    VideoCompress264SD

    H.264 SD transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0000707547

    per second of video

    VideoCompress264HD

    H.264 HD transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0001273585

    per second of video

    VideoCompress2642K

    H.264 2K transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0002830189

    per second of video

    VideoCompress2644K

    H.264 4K transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0006367925

    per second of video

    VideoCompress265LD

    H.265 LD transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0002122642

    per second of video

    VideoCompress265SD

    H.265 SD transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0003537736

    per second of video

    VideoCompress265HD

    H.265 HD transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0007075472

    per second of video

    VideoCompress2652K

    H.265 2K transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0011320755

    per second of video

    VideoCompress2654K

    H.265 4K transcodingNote*

    CreateMediaConvertTask (create a media transcoding task)

    • video/concat

    • video/convert

    0.0022641509

    per second of video

    MediaAnimation

    Video Animation

    CreateMediaConvertTask (create a media transcoding task)

    video/animation

    • Basic format: 0.012

    • Advanced format: 0.074

    Thousands of frames

    ExtractSubtitleText

    Text subtitle extraction from videos

    CreateMediaConvertTask (create a media transcoding task)

    0.223

    Qianlu

    ExtractSubtitleImage

    Image subtitle extraction from videos

    CreateMediaConvertTask (create a media transcoding task)

    0.015

    Thousand Frames

    VideoFraming

    Video snapshot

    CreateMediaConvertTask (create a media transcoding task)

    • video/snapshots

    • video/sprite

    0.015

    Thousand Frames

    VideoClassification

    Video label detection

    CreateVideoLabelClassificationTask (create a media transcoding task)

    7.0754717

    per 1,000 calls

    LiveTranscoding

    Real-time transcodingNote*

    GenerateVideoPlaylist (generate a real-time transcoding playlist)

    • hls/m3u8

    • hls/ts

    0.0000141509

    CountUnit

    Document Processing

    For detailed pricing of document processing billing items, refer to the following table.

    Important

    Projects created before December 1, 2023 for online preview and online editing are billed based on the number of times documents are opened. Projects created on or after December 1, 2023 are billed based on the number of API calls.

    Billing Item

    Billing Item Description

    Related API Operations

    Related x-oss-process Operations

    Price (USD)

    Unit

    DocumentConvert

    Document conversion

    CreateOfficeConversionTask (create a document conversion task)

    • doc/convert

    • doc/snapshot

    11.3207547

    per 1,000 calls

    Document content extraction

    ExtractDocumentText (extract document text)

    DocumentWebofficeEdit

    Online editing (Weboffice)Note*

    • GenerateWebofficeToken (obtain a Weboffice token)

    • RefreshWebofficeToken (refresh a Weboffice token)

    doc/edit

    2.8301887

    per 1,000 calls

    DocumentWebofficePreview

    Online preview (Weboffice)Note*

    • GenerateWebofficeToken (obtain a Weboffice token)

    • RefreshWebofficeToken (refresh a Weboffice token)

    doc/preview

    1.4150943

    per 1,000 calls

    DocumentWebofficeCachePreview

    Cached preview (Weboffice)

    • GenerateWebofficeToken (obtain a Weboffice token)

    • RefreshWebofficeToken (refresh a Weboffice token)

    0.9905660

    per 1,000 calls

    Important

    This refers to the number of API calls.

    File Processing

    For detailed pricing of file processing billing items, refer to the following table.

    Billing Item

    Billing Project Description

    Related API Operations

    Related x-oss-process Operations

    Price (USD)

    Unit

    PointCloudCompress

    Point cloud compression

    CreateCompressPointCloudTask (create a point cloud compression task)

    pointcloud/compress

    0.03

    per 1,000 calls

    FileProcess

    File packaging and download

    CreateFileCompressionTask (create a file compression task)

    0.00074

    GB

    Archive decompression

    CreateFileUncompressionTask (create a decompression task)

    FilePreview

    Archive preview

    CreateArchiveFileInspectionTask (create an archive preview and parsing task)

    0.0074

    TB

    API operations with multiple billing items

    Note

    • The SemanticQuery operation incurs charges for both StandardQueryL2 and SemanticAnalyze.

    • The CreateFacesSearchingTask operation incurs charges for both ImageDetect and ImageFace.

    Video transcoding details

    Note

    • H.264 transcoding: Output video uses the H.264 encoder.

    • H.265 transcoding: Output video uses the H.265 encoder.

    • LD: Output resolution is less than or equal to 640 × 480.

    • SD: Output resolution is less than or equal to 1280 × 720.

    • HD: Output resolution is less than or equal to 1920 × 1080.

    • 2K: Output resolution is less than or equal to 2560 × 1440.

    • 4K: Output resolution is less than or equal to 3840 × 2160.

    • Video transcoding is billed per second of output video. Charges are rounded up to the nearest second. Fractions of a second are billed as a full second.

    Document preview and editing billing details

    Note

    • For projects created before December 1, 2023, online editing and online preview are billed based on the number of times documents are opened—not the number of API calls.

    • Projects created on or after December 1, 2023 are billed based on the number of API calls. To switch to the new billing model, create a new project.

    • In the API call billing model, each API call can be used by only one user. If an API call is reused, only the last user retains access, and access permissions for other users are revoked.

    • If the Permission.Readonly parameter in the GenerateWebofficeToken operation is set to true, you are charged for document preview. If it is set to false, you are charged for online editing.

    • The RefreshWebofficeToken operation is billed based on the Permission.Readonly setting used when generating the original token. If Permission.Readonly was set to true, you are charged for document preview. Otherwise, you are charged for online editing.

    Real-time transcoding billing details

    Note

    • Billing components:

      • When generating a playlist, you can control initial transcoding duration using the InitialTranscode parameter. This incurs LiveTranscoding charges.

      • During playback, if a .ts file has not been transcoded, a new transcoding job starts. This incurs LiveTranscoding charges.

      • Reading source video files from OSS and writing transcoded files back to OSS incurs fees. Reading video files from OSS for playback also incurs fees. For details about OSS-related fees, refer to the OSS billing items.

    • LiveTranscoding count formula:

      • Video

        • Codec `eff` values: h264 = 0.3, h265 = 1.8

        • Formula:

      Ceiling(eff * Ceiling(Height/240) * Ceiling(Width/240) * Ceiling(FrameRate/30) + 1) * Ceiling(VideoStreamDuration)

      • Audio

        • `eff` value: 0.3

        • Formula:

      Ceiling(eff * Ceiling(AudioStreamDuration))

    • Billing rule: Each audio or video stream specified in TargetVideo.Stream or TargetAudio.Stream is processed separately and billed individually. The following examples illustrate real-time transcoding costs.

      • Example 1 (playlist generated only; no playback, so no LiveTranscoding charges):

        • A user calls GenerateVideoPlaylist. The output video length is 38 minutes, with a resolution of 800 × 600, a frame rate of 30, and an h264 video codec. The initial transcoding duration is 0 seconds, and TranscodeAhead uses the default value. No video is played.

      • Example 2 (playlist generated only; pre-transcoding configured; only pre-transcoding incurs LiveTranscoding charges):

        • A user calls GenerateVideoPlaylist. The output video length is 38 minutes, with a resolution of 800 × 600, a frame rate of 30, and an h264 video codec. The initial transcoding duration is 30 seconds, and TranscodeAhead uses the default value. No video is played.

        • Charges:

          • LiveTranscoding (CU count formula): Ceiling((0.3 * Ceiling(800/240) * Ceiling(600/240) * Ceiling(30/30) + 1) * Ceiling(30)) + Ceiling(0.3 * Ceiling(30)) = 159 (CountUnit)

      • Example 3 (playlist generated and partially played; only played segments incur LiveTranscoding charges):

        • A user calls GenerateVideoPlaylist. The output video length is 38 minutes, with a resolution of 800 × 600, a frame rate of 30, and an h264 video codec. The initial transcoding duration is 0 seconds, and TranscodeAhead uses the default value. The user plays the m3u8 playlist from the start, stops at minute 5 (default forward transcoding is 2 minutes), then jumps to minute 15 and plays to the end.

        • Charges:

          • LiveTranscoding (CU count formula): Ceiling((0.3 * Ceiling(800/240) * Ceiling(600/240) * Ceiling(30/30) + 1) * (Ceiling((5+2)*60) + Ceiling((38-15)*60))) + Ceiling(0.3 * Ceiling((5+2) * 60)) + Ceiling(0.3 * Ceiling((38-15) * 60)) = 9540 (CountUnit)

      • Example 4 (multiple users play the same video; repeated playback incurs LiveTranscoding charges only once):

        • A user calls GenerateVideoPlaylist. The output video length is 38 minutes, with a resolution of 800 × 600, a frame rate of 30, and an h264 video codec. The initial transcoding duration is 0 seconds, and TranscodeAhead uses the default value.

          User A plays the m3u8 playlist from the start to minute 5, then exits.

          User B plays the m3u8 playlist from minute 15 to the end.

          User C plays the m3u8 playlist from start to finish.

        • Charges:

          • LiveTranscoding (CU count formula): Ceiling((0.3 * Ceiling(800/240) * Ceiling(600/240) * Ceiling(30/30) + 1) * Ceiling(38*60)) + Ceiling(0.3 * Ceiling(38 * 60)) = 12084 (CountUnit)

    • Term definitions:

      • Width: The width of the output video resolution.

      • Height: The height of the output video resolution.

      • FrameRate: The video frame rate.

      • VideoStreamDuration: The duration of the video stream.

      • AudioStreamDuration: The length of the audio stream.

      • `eff`: The CU coefficient.

      • Ceiling(x) function: Returns the smallest integer greater than or equal to x.

    Operator and billing item mapping

    When you create metadata indexes by binding an OSS Bucket or calling IndexFileMeta or BatchIndexFileMeta, executing operators from the workflow template and operator mapping generates data processing fees, index storage fees, and OSS request fees. OSS request fees are charged by OSS. For more information, refer to OSS request fees. The mapping between operators and billing items is as follows:

    Operator

    Billing Item

    Billed by

    OSSMeta operator

    GetRequest

    OSS

    MIME operator

    No charge

    N/A

    FaceDetection operator

    ImageFaceNote*

    IMM

    LabelClassification operator (image)

    ImageClassificationNote*

    IMM

    LabelClassification operator (video)

    VideoClassification

    IMM

    ImageScoring operator

    ImageScoringNote*

    IMM

    ReGEO operator

    ReverseGeocoding

    IMM

    MediaMeta operator

    MediaMeta

    IMM

    EXIF operator

    GetRequest

    OSS

    ExtractDocumentText operator

    DocumentConvert

    IMM

    ExtractImageEmbeddings operator

    Free for a limited time

    IMM

    Important

    To process various image formats, IMM uses the image processing feature of Object Storage Service (OSS) to perform one or more operations, such as format conversion or image scaling. These operations incur fees charged by OSS. For more information about these fees, refer to Data processing fees.

    External request fees

    Accessing OSS through Intelligent Media Management (IMM) incurs OSS request fees. For more information, refer to Request fees.