Skip to content

📚 References

EmbedAnything: A high-performance, multimodal embedding pipeline.

This module provides functions and classes for embedding queries, files, and directories using different embedding models. It supports text, images, audio, videos, PDFs, and other media types with various embedding backends (Candle, ONNX, Cloud).

Main Functions:

  • embed_query: Embeds text queries and returns a list of EmbedData objects.
  • embed_file: Embeds a single file and returns a list of EmbedData objects.
  • embed_directory: Embeds all files in a directory and returns a list of EmbedData objects.
  • embed_image_directory: Embeds all images in a directory.
  • embed_audio_file: Embeds audio files using Whisper for transcription.
  • embed_video_file: Embeds a video file by sampling frames.
  • embed_video_directory: Embeds all videos in a directory.
  • embed_webpage: Embeds content from a webpage URL.

Main Classes:

  • EmbeddingModel: Main class for loading and using embedding models.
  • EmbedData: Represents embedded data with text, embedding vector, and metadata.
  • TextEmbedConfig: Configuration for text embedding (chunking, batching, etc.).
  • VideoEmbedConfig: Configuration for video embedding (frame sampling, batching).
  • ColpaliModel: Specialized model for document/image-text embedding.
  • ColbertModel: Model for late-interaction embeddings.
  • Reranker: Model for re-ranking search results.
  • AudioDecoderModel: Model for audio transcription (Whisper).

Usage Examples:

Text Embedding

from embed_anything import EmbeddingModel, WhichModel, TextEmbedConfig
import embed_anything

# Load a text embedding model
model = EmbeddingModel.from_pretrained_local(
    WhichModel.Bert, 
    model_id="sentence-transformers/all-MiniLM-L12-v2"
)

# Configure embedding parameters
config = TextEmbedConfig(
    chunk_size=1000,              # Characters per chunk
    batch_size=32,                # Process 32 chunks at once
    splitting_strategy="sentence"  # Split by sentences
)

# Embed a PDF file
data = embed_anything.embed_file("test_files/document.pdf", embedder=model, config=config)

# Access results
for item in data:
    print(f"Text: {item.text[:100]}...")
    print(f"Embedding dimension: {len(item.embedding)}")
    print(f"Metadata: {item.metadata}")

Image Embedding

import embed_anything
import numpy as np
from embed_anything import EmbedData, EmbeddingModel, WhichModel

# Load CLIP model for image-text embeddings
model = EmbeddingModel.from_pretrained_local(
    WhichModel.Clip,
    model_id="openai/clip-vit-base-patch16"
)

# Embed all images in a directory
data: list[EmbedData] = embed_anything.embed_image_directory(
    "test_files", 
    embedder=model
)

# Convert to numpy array for similarity search
embeddings = np.array([item.embedding for item in data])

# Embed a text query
query = ["Photo of a monkey?"]
query_embedding = np.array(
    embed_anything.embed_query(query, embedder=model)[0].embedding
)

# Calculate cosine similarity
similarities = np.dot(embeddings, query_embedding)
most_similar_idx = np.argmax(similarities)
print(f"Most similar image: {data[most_similar_idx].text}")

Audio Embedding

from embed_anything import (
    AudioDecoderModel,
    EmbeddingModel,
    embed_audio_file,
    TextEmbedConfig,
    WhichModel
)
import embed_anything

# Load Whisper model for audio transcription
# Choose from: https://huggingface.co/distil-whisper or 
# https://huggingface.co/collections/openai/whisper-release-6501bba2cf999715fd953013
audio_decoder = AudioDecoderModel.from_pretrained_hf(
    "openai/whisper-tiny.en", 
    revision="main", 
    model_type="tiny-en", 
    quantized=False
)

# Load text embedding model for transcribed text
embedder = EmbeddingModel.from_pretrained_hf(
    WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)

# Configure text embedding
config = TextEmbedConfig(chunk_size=200, batch_size=32)

# Embed audio file (transcribes then embeds)
data = embed_anything.embed_audio_file(
    "test_files/audio/samples_hp0.wav",
    audio_decoder=audio_decoder,
    embedder=embedder,
    text_embed_config=config,
)

# Access transcribed and embedded segments
for item in data:
    print(f"Transcribed text: {item.text}")
    print(f"Metadata: {item.metadata}")

Vector Database Integration

Store embeddings directly to a vector database without keeping them in memory:

import embed_anything
import os
from embed_anything import EmbeddingModel, WhichModel, TextEmbedConfig
from embed_anything.vectordb import PineconeAdapter

# Initialize Pinecone adapter
api_key = os.environ.get("PINECONE_API_KEY")
pinecone_adapter = PineconeAdapter(api_key)

# Create or use existing index
try:
    pinecone_adapter.delete_index("my-index")
except:
    pass

pinecone_adapter.create_index(
    dimension=512,      # Embedding dimension
    metric="cosine",    # Similarity metric
    index_name="my-index"
)

# Load embedding model
model = EmbeddingModel.from_pretrained_hf(
    WhichModel.Clip, 
    "openai/clip-vit-base-patch16"
)

# Embed images and stream directly to Pinecone
data = embed_anything.embed_image_directory(
    "test_files",
    embedder=model,
    adapter=pinecone_adapter,  # Streams to database
)

# Embeddings are now in Pinecone, not in memory
print("Embeddings stored in Pinecone!")

ONNX Models (Faster Inference)

from embed_anything import EmbeddingModel, WhichModel, ONNXModel, Dtype

# Load a pre-configured ONNX model (faster, lower memory)
model = EmbeddingModel.from_pretrained_onnx(
    WhichModel.Bert,
    model_id=ONNXModel.BGESmallENV15Q,  # Quantized BGE model
    dtype=Dtype.Q4F16
)

# Use like any other model
data = embed_anything.embed_file("test_files/document.pdf", embedder=model)

Semantic Chunking

from embed_anything import EmbeddingModel, WhichModel, TextEmbedConfig
import embed_anything

# Main embedding model
model = EmbeddingModel.from_pretrained_hf(
    WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L12-v2"
)

# Semantic encoder for chunk boundaries
semantic_encoder = EmbeddingModel.from_pretrained_hf(
    WhichModel.Jina,
    model_id="jinaai/jina-embeddings-v2-small-en"
)

# Configure semantic chunking
config = TextEmbedConfig(
    chunk_size=1000,
    batch_size=32,
    splitting_strategy="semantic",
    semantic_encoder=semantic_encoder
)

# Embed with semantic chunking
data = embed_anything.embed_file("test_files/document.pdf", embedder=model, config=config)

Supported Embedding Models:

  • Text Models: BERT, Jina, Qwen3, Splade, ColBERT, Model2Vec
  • Image Models: CLIP, SigLip
  • Audio Models: Whisper, DistilWhisper
  • Document Models: ColPali
  • Rerankers: Jina Reranker, BGE Reranker, Qwen3 Reranker
  • Cloud Models: OpenAI, Cohere, Gemini

For more examples and detailed documentation, visit: https://embed-anything.com

Adapter

Bases: ABC

Source code in python/python/embed_anything/_embed_anything.pyi 
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
class Adapter(ABC):
    def __init__(self, api_key: str):
        """
        Initializes the Adapter object.

        Args:
            api_key: The API key for accessing the adapter.
        """

    @abstractmethod
    def create_index(self, dimension: int, metric: str, index_name: str, **kwargs): ...
    """
    Creates an index for storing the embeddings.

    Args:
        dimension: The dimension of the embeddings.
        metric: The metric for measuring the distance between embeddings.
        index_name: The name of the index.
        kwargs: Additional keyword arguments.
    """
    @abstractmethod
    def delete_index(self, index_name: str):
        """
        Deletes an index.

        Args:
            index_name: The name of the index to delete.
        """

    @abstractmethod
    def convert(self, embeddings: List[List[EmbedData]]) -> List[Dict]:
        """
        Converts the embeddings to a list of dictionaries.

        Args:
            embeddings: The list of embeddings.

        Returns:
            A list of dictionaries.
        """

    @abstractmethod
    def upsert(self, data: List[Dict]):
        """
        Upserts the data into the index.

        Args:
            data: The list of data to upsert.
        """

__init__(api_key)

Initializes the Adapter object.

Parameters:

Name Type Description Default
api_key str

The API key for accessing the adapter.

 required
Source code in python/python/embed_anything/_embed_anything.pyi 
 6
 7
 8
 9
10
11
12
def __init__(self, api_key: str):
    """
    Initializes the Adapter object.

    Args:
        api_key: The API key for accessing the adapter.
    """

convert(embeddings) abstractmethod

Converts the embeddings to a list of dictionaries.

Parameters:

Name Type Description Default
embeddings List[List[EmbedData]]

The list of embeddings.

 required

Returns:

Type Description
List[Dict]

A list of dictionaries.
Source code in python/python/embed_anything/_embed_anything.pyi 
34
35
36
37
38
39
40
41
42
43
44
@abstractmethod
def convert(self, embeddings: List[List[EmbedData]]) -> List[Dict]:
    """
    Converts the embeddings to a list of dictionaries.

    Args:
        embeddings: The list of embeddings.

    Returns:
        A list of dictionaries.
    """

delete_index(index_name) abstractmethod

Deletes an index.

Parameters:

Name Type Description Default
index_name str

The name of the index to delete.

 required
Source code in python/python/embed_anything/_embed_anything.pyi 
25
26
27
28
29
30
31
32
@abstractmethod
def delete_index(self, index_name: str):
    """
    Deletes an index.

    Args:
        index_name: The name of the index to delete.
    """

upsert(data) abstractmethod

Upserts the data into the index.

Parameters:

Name Type Description Default
data List[Dict]

The list of data to upsert.

 required
Source code in python/python/embed_anything/_embed_anything.pyi 
46
47
48
49
50
51
52
53
@abstractmethod
def upsert(self, data: List[Dict]):
    """
    Upserts the data into the index.

    Args:
        data: The list of data to upsert.
    """

AudioDecoderModel

Represents an audio decoder model.

Attributes:

Name Type Description
model_id str

The ID of the audio decoder model.
revision str

The revision of the audio decoder model.
model_type str

The type of the audio decoder model.
quantized bool

A flag indicating whether the audio decoder model is quantized or not.

Example: 

model = embed_anything.AudioDecoderModel.from_pretrained_hf(
    model_id="openai/whisper-tiny.en",
    revision="main",
    model_type="tiny-en",
    quantized=False
)

Source code in python/python/embed_anything/_embed_anything.pyi 
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
class AudioDecoderModel:
    """
    Represents an audio decoder model.

    Attributes:
        model_id: The ID of the audio decoder model.
        revision: The revision of the audio decoder model.
        model_type: The type of the audio decoder model.
        quantized: A flag indicating whether the audio decoder model is quantized or not.

    Example:
    ```python

    model = embed_anything.AudioDecoderModel.from_pretrained_hf(
        model_id="openai/whisper-tiny.en",
        revision="main",
        model_type="tiny-en",
        quantized=False
    )
    ```
    """

    model_id: str
    revision: str
    model_type: str
    quantized: bool

    def from_pretrained_hf(
        model_id: str | None = None,
        revision: str | None = None,
        model_type: str | None = None,
        quantized: bool | None = None,
    ): ...

ColbertModel

Represents the Colbert model.

Source code in python/python/embed_anything/_embed_anything.pyi 
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
class ColbertModel:
    """
    Represents the Colbert model.
    """

    def __init__(
        self,
        hf_model_id: str | None = None,
        revision: str | None = None,
        path_in_repo: str | None = None,
    ):
        """
        Initializes the ColbertModel object.
        """

    def from_pretrained_onnx(
        self,
        hf_model_id: str | None = None,
        revision: str | None = None,
        path_in_repo: str | None = None,
    ) -> ColbertModel:
        """
        Loads a pre-trained Colbert model from the Hugging Face model hub.

        Attributes:
            hf_model_id: The ID of the model from Hugging Face.
            revision: The revision of the model.
            path_in_repo: The path to the model in the repository.

        Returns:
            A ColbertModel object.
        """

    def embed(
        self, text_batch: list[str], batch_size: int | None = None, is_doc: bool = True
    ) -> list[EmbedData]:
        """
        Embeds the given text and returns a list of EmbedData objects.
        """

__init__(hf_model_id=None, revision=None, path_in_repo=None)

Initializes the ColbertModel object.

Source code in python/python/embed_anything/_embed_anything.pyi 
428
429
430
431
432
433
434
435
436
def __init__(
    self,
    hf_model_id: str | None = None,
    revision: str | None = None,
    path_in_repo: str | None = None,
):
    """
    Initializes the ColbertModel object.
    """

embed(text_batch, batch_size=None, is_doc=True)

Embeds the given text and returns a list of EmbedData objects.

Source code in python/python/embed_anything/_embed_anything.pyi 
456
457
458
459
460
461
def embed(
    self, text_batch: list[str], batch_size: int | None = None, is_doc: bool = True
) -> list[EmbedData]:
    """
    Embeds the given text and returns a list of EmbedData objects.
    """

from_pretrained_onnx(hf_model_id=None, revision=None, path_in_repo=None)

Loads a pre-trained Colbert model from the Hugging Face model hub.

Attributes:

Name Type Description
hf_model_id

The ID of the model from Hugging Face.
revision

The revision of the model.
path_in_repo

The path to the model in the repository.

Returns:

Type Description
ColbertModel

A ColbertModel object.
Source code in python/python/embed_anything/_embed_anything.pyi 
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
def from_pretrained_onnx(
    self,
    hf_model_id: str | None = None,
    revision: str | None = None,
    path_in_repo: str | None = None,
) -> ColbertModel:
    """
    Loads a pre-trained Colbert model from the Hugging Face model hub.

    Attributes:
        hf_model_id: The ID of the model from Hugging Face.
        revision: The revision of the model.
        path_in_repo: The path to the model in the repository.

    Returns:
        A ColbertModel object.
    """

ColpaliModel

Represents the Colpali model.

Source code in python/python/embed_anything/_embed_anything.pyi 
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
class ColpaliModel:
    """
    Represents the Colpali model.
    """

    def __init__(self, model_id: str, revision: str | None = None):
        """
        Initializes the ColpaliModel object.

        Args:
            model_id: The ID of the model from Hugging Face.
            revision: The revision of the model.
        """

    def from_pretrained(model_id: str, revision: str | None = None) -> ColpaliModel:
        """
        Loads a pre-trained Colpali model from the Hugging Face model hub.

        Args:
            model_id: The ID of the model from Hugging Face.
            revision: The revision of the model.

        Returns:
            A ColpaliModel object.
        """

    def from_pretrained_onnx(
        model_id: str, revision: str | None = None
    ) -> ColpaliModel:
        """
        Loads a pre-trained Colpali model from the Hugging Face model hub.

        Args:
            model_id: The ID of the model from Hugging Face.
            revision: The revision of the model.

        Returns:
            A ColpaliModel object.
        """

    def embed_file(self, file_path: str, batch_size: int | None = 1) -> list[EmbedData]:
        """
        Embeds the given pdf file and returns a list of EmbedData objects for each page in the file This first convert the pdf file into images and then embed each image.

        Args:
            file_path: The path to the pdf file to embed.
            batch_size: The batch size for processing the embeddings. Default is 1.

        Returns:
            A list of EmbedData objects for each page in the file.
        """

    def embed_query(self, query: str) -> list[EmbedData]:
        """
        Embeds the given query and returns a list of EmbedData objects.

        Args:
            query: The query to embed.

        Returns:
            A list of EmbedData objects.

        """

__init__(model_id, revision=None)

Initializes the ColpaliModel object.

Parameters:

Name Type Description Default
model_id str

The ID of the model from Hugging Face.

 required
revision str | None

The revision of the model.

 None
Source code in python/python/embed_anything/_embed_anything.pyi 
364
365
366
367
368
369
370
371
def __init__(self, model_id: str, revision: str | None = None):
    """
    Initializes the ColpaliModel object.

    Args:
        model_id: The ID of the model from Hugging Face.
        revision: The revision of the model.
    """

embed_file(file_path, batch_size=1)

Embeds the given pdf file and returns a list of EmbedData objects for each page in the file This first convert the pdf file into images and then embed each image.

Parameters:

Name Type Description Default
file_path str

The path to the pdf file to embed.

 required
batch_size int | None

The batch size for processing the embeddings. Default is 1.

 1

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects for each page in the file.
Source code in python/python/embed_anything/_embed_anything.pyi 
399
400
401
402
403
404
405
406
407
408
409
def embed_file(self, file_path: str, batch_size: int | None = 1) -> list[EmbedData]:
    """
    Embeds the given pdf file and returns a list of EmbedData objects for each page in the file This first convert the pdf file into images and then embed each image.

    Args:
        file_path: The path to the pdf file to embed.
        batch_size: The batch size for processing the embeddings. Default is 1.

    Returns:
        A list of EmbedData objects for each page in the file.
    """

embed_query(query)

Embeds the given query and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
query str

The query to embed.

 required

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
411
412
413
414
415
416
417
418
419
420
421
def embed_query(self, query: str) -> list[EmbedData]:
    """
    Embeds the given query and returns a list of EmbedData objects.

    Args:
        query: The query to embed.

    Returns:
        A list of EmbedData objects.

    """

from_pretrained(model_id, revision=None)

Loads a pre-trained Colpali model from the Hugging Face model hub.

Parameters:

Name Type Description Default
model_id str

The ID of the model from Hugging Face.

 required
revision str | None

The revision of the model.

 None

Returns:

Type Description
ColpaliModel

A ColpaliModel object.
Source code in python/python/embed_anything/_embed_anything.pyi 
373
374
375
376
377
378
379
380
381
382
383
def from_pretrained(model_id: str, revision: str | None = None) -> ColpaliModel:
    """
    Loads a pre-trained Colpali model from the Hugging Face model hub.

    Args:
        model_id: The ID of the model from Hugging Face.
        revision: The revision of the model.

    Returns:
        A ColpaliModel object.
    """

from_pretrained_onnx(model_id, revision=None)

Loads a pre-trained Colpali model from the Hugging Face model hub.

Parameters:

Name Type Description Default
model_id str

The ID of the model from Hugging Face.

 required
revision str | None

The revision of the model.

 None

Returns:

Type Description
ColpaliModel

A ColpaliModel object.
Source code in python/python/embed_anything/_embed_anything.pyi 
385
386
387
388
389
390
391
392
393
394
395
396
397
def from_pretrained_onnx(
    model_id: str, revision: str | None = None
) -> ColpaliModel:
    """
    Loads a pre-trained Colpali model from the Hugging Face model hub.

    Args:
        model_id: The ID of the model from Hugging Face.
        revision: The revision of the model.

    Returns:
        A ColpaliModel object.
    """

DocumentRank

Represents the rank of a document.

Attributes:

Name Type Description
document str

The document to rank.
relevance_score float

The relevance score of the document.
rank int

The rank of the document.
Source code in python/python/embed_anything/_embed_anything.pyi 
544
545
546
547
548
549
550
551
552
553
554
555
556
class DocumentRank:
    """
    Represents the rank of a document.

    Attributes:
        document: The document to rank.
        relevance_score: The relevance score of the document.
        rank: The rank of the document.
    """

    document: str
    relevance_score: float
    rank: int

Dtype

Bases: Enum

Represents the data type of the model.

Source code in python/python/embed_anything/_embed_anything.pyi 
518
519
520
521
522
523
524
525
526
527
528
529
530
class Dtype(Enum):
    """
    Represents the data type of the model.
    """

    F16 = "F16"
    INT8 = "INT8"
    Q4 = "Q4"
    UINT8 = "UINT8"
    BNB4 = "BNB4"
    Q4F16 = "Q4F16"
    BF16 = "BF16"
    F32 = "F32"

EmbedData

Represents the data of an embedded file.

Attributes:

Name Type Description
embedding list[float]

The embedding of the file.
text str

The text for which the embedding is generated for.
metadata dict[str, str]

Additional metadata associated with the embedding.
Source code in python/python/embed_anything/_embed_anything.pyi 
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
class EmbedData:
    """Represents the data of an embedded file.

    Attributes:
        embedding: The embedding of the file.
        text: The text for which the embedding is generated for.
        metadata: Additional metadata associated with the embedding.
    """

    def __init__(self, embedding: list[float], text: str, metadata: dict[str, str]):
        self.embedding = embedding
        self.text = text
        self.metadata = metadata
    embedding: list[float]
    text: str
    metadata: dict[str, str]

EmbeddingModel

Represents an embedding model.

Source code in python/python/embed_anything/_embed_anything.pyi 
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
class EmbeddingModel:
    """
    Represents an embedding model.
    """

    def from_pretrained_hf(
        model_id: str,
        revision: str | None = None,
        token: str | None = None,
        dtype: Dtype | None = None,
        pooling: Pooling | None = None,
    ) -> EmbeddingModel:
        """
        Loads an embedding model from the Hugging Face model hub.

        The model architecture (BERT, Jina, CLIP, ModernBERT, Qwen3, Gemma3,
        Model2Vec, SPLADE, ColPali, etc.) is auto-detected from the model's
        `config.json`, so `WhichModel` is no longer required here.

        Attributes:
            model_id: The ID of the model on the Hugging Face Hub.
            revision: The revision (branch, tag, or commit) of the model.
            token: The Hugging Face access token. Required for private or gated
                repositories. If None, the token is read from the `HF_TOKEN`
                environment variable (or the local Hugging Face login) when available.
            dtype: The dtype of the model.
            pooling: The pooling strategy used to turn token embeddings into a
                single sentence embedding (`Pooling.Mean`, `Pooling.Cls`, or
                `Pooling.LastToken`). If None, the model's default pooling is used.
        Returns:
            An EmbeddingModel object.

        Example:
        ```python
        from embed_anything import EmbeddingModel, Pooling

        model = EmbeddingModel.from_pretrained_hf(
            model_id="sentence-transformers/all-MiniLM-L6-v2",
            revision="main",
            pooling=Pooling.Mean,
        )

        # Loading a private or gated model with an access token
        gated_model = EmbeddingModel.from_pretrained_hf(
            model_id="google/embeddinggemma-300m",
            token="hf_...",
        )
        ```

        """

    def from_pretrained_cloud(
        model: WhichModel, model_id: str, api_key: str | None = None
    ) -> EmbeddingModel:
        """
        Loads an embedding model from a cloud-based service.

        Attributes:
            model (WhichModel): The cloud service to use. Currently supports WhichModel.OpenAI and WhichModel.Cohere.
            model_id (str): The ID of the model to use.

                - For OpenAI, see available models at https://platform.openai.com/docs/guides/embeddings/embedding-models
                - For Cohere, see available models at https://docs.cohere.com/docs/cohere-embed
                - For CohereVision, see available models at https://docs.cohere.com/docs/cohere-embed
            api_key (str | None, optional): The API key for accessing the model. If not provided, it is taken from the environment variable:

                - For OpenAI: OPENAI_API_KEY
                - For Cohere: CO_API_KEY

        Returns:
            EmbeddingModel: An initialized EmbeddingModel object.

        Raises:
            ValueError: If an unsupported model is specified.

        Example:
        ```python
        # Using Cohere
        model = EmbeddingModel.from_pretrained_cloud(
            model=WhichModel.Cohere,
            model_id="embed-english-v3.0"
        )

        # Using OpenAI
        model = EmbeddingModel.from_pretrained_cloud(
            model=WhichModel.OpenAI,
            model_id="text-embedding-3-small"
        )
        ```
        """

    def from_pretrained_onnx(
        model: WhichModel,
        model_name: Optional[ONNXModel] | None = None,
        hf_model_id: Optional[str] | None = None,
        revision: Optional[str] | None = None,
        dtype: Optional[Dtype] | None = None,
        path_in_repo: Optional[str] | None = None,
    ) -> EmbeddingModel:
        """
        Loads an ONNX embedding model.

        Args:
            model (WhichModel): The architecture of the embedding model to use.
            model_name (ONNXModel | None, optional): The name of the model. Defaults to None.
            hf_model_id (str | None, optional): The ID of the model from Hugging Face. Defaults to None.
            revision (str | None, optional): The revision of the model. Defaults to None.
            dtype (Dtype | None, optional): The dtype of the model. Defaults to None.
            path_in_repo (str | None, optional): The path to the model in the repository. Defaults to None.
        Returns:
            EmbeddingModel: An initialized EmbeddingModel object.

        Atleast one of the following arguments must be provided:
            - model_name
            - hf_model_id

        If hf_model_id is provided, dtype is ignored and the path_in_repo has to be provided pointing to the model file in the repository.
        If model_name is provided, dtype is used to determine the model file to load.

        Example:
        ```python
        model = EmbeddingModel.from_pretrained_onnx(
            model=WhichModel.Bert,
            model_name=ONNXModel.BGESmallENV15Q,
            dtype=Dtype.Q4F16
        )

        model = EmbeddingModel.from_pretrained_onnx(
            model=WhichModel.Bert,
            hf_model_id="jinaai/jina-embeddings-v3",
            path_in_repo="onnx/model_fp16.onnx"
        )
        ```

        Note:
        This method loads a pre-trained model in ONNX format, which can offer improved inference speed
        compared to standard PyTorch models. ONNX models are particularly useful for deployment
        scenarios where performance is critical.
        """

    def embed_file(
        self,
        file_path: str,
        config: TextEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given file and returns a list of EmbedData objects.

        Args:
            file_path: The path to the file to embed.
            config: The configuration for the embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_files_batch(
        self,
        files: list[str],
        config: TextEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given files and returns a list of EmbedData objects.

        Args:
            files: The list of files to embed.
            config: The configuration for the embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_audio_file(
        self,
        audio_file: str,
        audio_decoder: AudioDecoderModel,
        config: TextEmbedConfig | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given audio file and returns a list of EmbedData objects.

        Args:
            audio_file: The path to the audio file to embed.
            audio_decoder: The audio decoder for the audio file.
            config: The configuration for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_video_file(
        self,
        video_file: str,
        config: VideoEmbedConfig | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given video file and returns a list of EmbedData objects.

        Args:
            video_file: The path to the video file to embed.
            config: The configuration for video embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_video_directory(
        self,
        directory: str,
        config: VideoEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData] | None:
        """
        Embeds videos in the given directory and returns a list of EmbedData objects.

        Args:
            directory: The path to the directory to embed.
            config: The configuration for video embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects, or None if an adapter is used.
        """

    def embed_query(
        self,
        query: list[str],
        config: TextEmbedConfig | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given list of queries and returns a list of EmbedData objects.

        Args:
            query: The list of queries to embed.
            config: The configuration for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_webpage(
        self,
        url: str,
        config: TextEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given webpage and returns a list of EmbedData objects.

        Args:
            url: The URL of the webpage to embed.
            config: The configuration for the embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_directory(
        self,
        directory: str,
        config: TextEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given directory and returns a list of EmbedData objects.

        Args:
            directory: The path to the directory to embed.
            config: The configuration for the embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_directory_stream(
        self,
        directory: str,
        config: TextEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given directory and returns a list of EmbedData objects.

        Args:
            directory: The path to the directory to embed.
            config: The configuration for the embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects.
        """

    def embed_webpage(
        self,
        url: str,
        config: TextEmbedConfig | None = None,
        adapter: Adapter | None = None,
    ) -> list[EmbedData]:
        """
        Embeds the given webpage and returns a list of EmbedData objects.

        Args:
            url: The URL of the webpage to embed.
            config: The configuration for the embedding.
            adapter: The adapter for the embedding.

        Returns:
            A list of EmbedData objects.
        """

embed_audio_file(audio_file, audio_decoder, config=None)

Embeds the given audio file and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
audio_file str

The path to the audio file to embed.

 required
audio_decoder AudioDecoderModel

The audio decoder for the audio file.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
def embed_audio_file(
    self,
    audio_file: str,
    audio_decoder: AudioDecoderModel,
    config: TextEmbedConfig | None = None,
) -> list[EmbedData]:
    """
    Embeds the given audio file and returns a list of EmbedData objects.

    Args:
        audio_file: The path to the audio file to embed.
        audio_decoder: The audio decoder for the audio file.
        config: The configuration for the embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_directory(directory, config=None, adapter=None)

Embeds the given directory and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
directory str

The path to the directory to embed.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None
adapter Adapter | None

The adapter for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
def embed_directory(
    self,
    directory: str,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given directory and returns a list of EmbedData objects.

    Args:
        directory: The path to the directory to embed.
        config: The configuration for the embedding.
        adapter: The adapter for the embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_directory_stream(directory, config=None, adapter=None)

Embeds the given directory and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
directory str

The path to the directory to embed.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None
adapter Adapter | None

The adapter for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
def embed_directory_stream(
    self,
    directory: str,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given directory and returns a list of EmbedData objects.

    Args:
        directory: The path to the directory to embed.
        config: The configuration for the embedding.
        adapter: The adapter for the embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_file(file_path, config=None, adapter=None)

Embeds the given file and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the file to embed.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None
adapter Adapter | None

The adapter for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
def embed_file(
    self,
    file_path: str,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given file and returns a list of EmbedData objects.

    Args:
        file_path: The path to the file to embed.
        config: The configuration for the embedding.
        adapter: The adapter for the embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_files_batch(files, config=None, adapter=None)

Embeds the given files and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
files list[str]

The list of files to embed.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None
adapter Adapter | None

The adapter for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
def embed_files_batch(
    self,
    files: list[str],
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given files and returns a list of EmbedData objects.

    Args:
        files: The list of files to embed.
        config: The configuration for the embedding.
        adapter: The adapter for the embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_query(query, config=None)

Embeds the given list of queries and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
query list[str]

The list of queries to embed.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
def embed_query(
    self,
    query: list[str],
    config: TextEmbedConfig | None = None,
) -> list[EmbedData]:
    """
    Embeds the given list of queries and returns a list of EmbedData objects.

    Args:
        query: The list of queries to embed.
        config: The configuration for the embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_video_directory(directory, config=None, adapter=None)

Embeds videos in the given directory and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
directory str

The path to the directory to embed.

 required
config VideoEmbedConfig | None

The configuration for video embedding.

 None
adapter Adapter | None

The adapter for the embedding.

 None

Returns:

Type Description
list[EmbedData] | None

A list of EmbedData objects, or None if an adapter is used.
Source code in python/python/embed_anything/_embed_anything.pyi 
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
def embed_video_directory(
    self,
    directory: str,
    config: VideoEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData] | None:
    """
    Embeds videos in the given directory and returns a list of EmbedData objects.

    Args:
        directory: The path to the directory to embed.
        config: The configuration for video embedding.
        adapter: The adapter for the embedding.

    Returns:
        A list of EmbedData objects, or None if an adapter is used.
    """

embed_video_file(video_file, config=None)

Embeds the given video file and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
video_file str

The path to the video file to embed.

 required
config VideoEmbedConfig | None

The configuration for video embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
def embed_video_file(
    self,
    video_file: str,
    config: VideoEmbedConfig | None = None,
) -> list[EmbedData]:
    """
    Embeds the given video file and returns a list of EmbedData objects.

    Args:
        video_file: The path to the video file to embed.
        config: The configuration for video embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_webpage(url, config=None, adapter=None)

Embeds the given webpage and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
url str

The URL of the webpage to embed.

 required
config TextEmbedConfig | None

The configuration for the embedding.

 None
adapter Adapter | None

The adapter for the embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
def embed_webpage(
    self,
    url: str,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given webpage and returns a list of EmbedData objects.

    Args:
        url: The URL of the webpage to embed.
        config: The configuration for the embedding.
        adapter: The adapter for the embedding.

    Returns:
        A list of EmbedData objects.
    """

from_pretrained_cloud(model, model_id, api_key=None)

Loads an embedding model from a cloud-based service.

Attributes:

Name Type Description
model WhichModel

The cloud service to use. Currently supports WhichModel.OpenAI and WhichModel.Cohere.
model_id str

The ID of the model to use.

  • For OpenAI, see available models at https://platform.openai.com/docs/guides/embeddings/embedding-models
  • For Cohere, see available models at https://docs.cohere.com/docs/cohere-embed
  • For CohereVision, see available models at https://docs.cohere.com/docs/cohere-embed
api_key str | None

The API key for accessing the model. If not provided, it is taken from the environment variable:

  • For OpenAI: OPENAI_API_KEY
  • For Cohere: CO_API_KEY

Returns:

Name Type Description
EmbeddingModel EmbeddingModel

An initialized EmbeddingModel object.

Raises:

Type Description
ValueError

If an unsupported model is specified.

Example: 

# Using Cohere
model = EmbeddingModel.from_pretrained_cloud(
    model=WhichModel.Cohere,
    model_id="embed-english-v3.0"
)

# Using OpenAI
model = EmbeddingModel.from_pretrained_cloud(
    model=WhichModel.OpenAI,
    model_id="text-embedding-3-small"
)

Source code in python/python/embed_anything/_embed_anything.pyi 
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
def from_pretrained_cloud(
    model: WhichModel, model_id: str, api_key: str | None = None
) -> EmbeddingModel:
    """
    Loads an embedding model from a cloud-based service.

    Attributes:
        model (WhichModel): The cloud service to use. Currently supports WhichModel.OpenAI and WhichModel.Cohere.
        model_id (str): The ID of the model to use.

            - For OpenAI, see available models at https://platform.openai.com/docs/guides/embeddings/embedding-models
            - For Cohere, see available models at https://docs.cohere.com/docs/cohere-embed
            - For CohereVision, see available models at https://docs.cohere.com/docs/cohere-embed
        api_key (str | None, optional): The API key for accessing the model. If not provided, it is taken from the environment variable:

            - For OpenAI: OPENAI_API_KEY
            - For Cohere: CO_API_KEY

    Returns:
        EmbeddingModel: An initialized EmbeddingModel object.

    Raises:
        ValueError: If an unsupported model is specified.

    Example:
    ```python
    # Using Cohere
    model = EmbeddingModel.from_pretrained_cloud(
        model=WhichModel.Cohere,
        model_id="embed-english-v3.0"
    )

    # Using OpenAI
    model = EmbeddingModel.from_pretrained_cloud(
        model=WhichModel.OpenAI,
        model_id="text-embedding-3-small"
    )
    ```
    """

from_pretrained_hf(model_id, revision=None, token=None, dtype=None, pooling=None)

Loads an embedding model from the Hugging Face model hub.

The model architecture (BERT, Jina, CLIP, ModernBERT, Qwen3, Gemma3, Model2Vec, SPLADE, ColPali, etc.) is auto-detected from the model's config.json, so WhichModel is no longer required here.

Attributes:

Name Type Description
model_id

The ID of the model on the Hugging Face Hub.
revision

The revision (branch, tag, or commit) of the model.
token

The Hugging Face access token. Required for private or gated repositories. If None, the token is read from the HF_TOKEN environment variable (or the local Hugging Face login) when available.
dtype

The dtype of the model.
pooling

The pooling strategy used to turn token embeddings into a single sentence embedding (Pooling.Mean, Pooling.Cls, or Pooling.LastToken). If None, the model's default pooling is used.

Returns: An EmbeddingModel object.

Example: 

from embed_anything import EmbeddingModel, Pooling

model = EmbeddingModel.from_pretrained_hf(
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
    pooling=Pooling.Mean,
)

# Loading a private or gated model with an access token
gated_model = EmbeddingModel.from_pretrained_hf(
    model_id="google/embeddinggemma-300m",
    token="hf_...",
)

Source code in python/python/embed_anything/_embed_anything.pyi 
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
def from_pretrained_hf(
    model_id: str,
    revision: str | None = None,
    token: str | None = None,
    dtype: Dtype | None = None,
    pooling: Pooling | None = None,
) -> EmbeddingModel:
    """
    Loads an embedding model from the Hugging Face model hub.

    The model architecture (BERT, Jina, CLIP, ModernBERT, Qwen3, Gemma3,
    Model2Vec, SPLADE, ColPali, etc.) is auto-detected from the model's
    `config.json`, so `WhichModel` is no longer required here.

    Attributes:
        model_id: The ID of the model on the Hugging Face Hub.
        revision: The revision (branch, tag, or commit) of the model.
        token: The Hugging Face access token. Required for private or gated
            repositories. If None, the token is read from the `HF_TOKEN`
            environment variable (or the local Hugging Face login) when available.
        dtype: The dtype of the model.
        pooling: The pooling strategy used to turn token embeddings into a
            single sentence embedding (`Pooling.Mean`, `Pooling.Cls`, or
            `Pooling.LastToken`). If None, the model's default pooling is used.
    Returns:
        An EmbeddingModel object.

    Example:
    ```python
    from embed_anything import EmbeddingModel, Pooling

    model = EmbeddingModel.from_pretrained_hf(
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
        pooling=Pooling.Mean,
    )

    # Loading a private or gated model with an access token
    gated_model = EmbeddingModel.from_pretrained_hf(
        model_id="google/embeddinggemma-300m",
        token="hf_...",
    )
    ```

    """

from_pretrained_onnx(model, model_name=None, hf_model_id=None, revision=None, dtype=None, path_in_repo=None)

Loads an ONNX embedding model.

Parameters:

Name Type Description Default
model WhichModel

The architecture of the embedding model to use.

 required
model_name ONNXModel | None

The name of the model. Defaults to None.

 None
hf_model_id str | None

The ID of the model from Hugging Face. Defaults to None.

 None
revision str | None

The revision of the model. Defaults to None.

 None
dtype Dtype | None

The dtype of the model. Defaults to None.

 None
path_in_repo str | None

The path to the model in the repository. Defaults to None.

 None

Returns: EmbeddingModel: An initialized EmbeddingModel object.

Atleast one of the following arguments must be provided
  • model_name
  • hf_model_id

If hf_model_id is provided, dtype is ignored and the path_in_repo has to be provided pointing to the model file in the repository. If model_name is provided, dtype is used to determine the model file to load.

Example: 

model = EmbeddingModel.from_pretrained_onnx(
    model=WhichModel.Bert,
    model_name=ONNXModel.BGESmallENV15Q,
    dtype=Dtype.Q4F16
)

model = EmbeddingModel.from_pretrained_onnx(
    model=WhichModel.Bert,
    hf_model_id="jinaai/jina-embeddings-v3",
    path_in_repo="onnx/model_fp16.onnx"
)

Note: This method loads a pre-trained model in ONNX format, which can offer improved inference speed compared to standard PyTorch models. ONNX models are particularly useful for deployment scenarios where performance is critical.

Source code in python/python/embed_anything/_embed_anything.pyi 
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
def from_pretrained_onnx(
    model: WhichModel,
    model_name: Optional[ONNXModel] | None = None,
    hf_model_id: Optional[str] | None = None,
    revision: Optional[str] | None = None,
    dtype: Optional[Dtype] | None = None,
    path_in_repo: Optional[str] | None = None,
) -> EmbeddingModel:
    """
    Loads an ONNX embedding model.

    Args:
        model (WhichModel): The architecture of the embedding model to use.
        model_name (ONNXModel | None, optional): The name of the model. Defaults to None.
        hf_model_id (str | None, optional): The ID of the model from Hugging Face. Defaults to None.
        revision (str | None, optional): The revision of the model. Defaults to None.
        dtype (Dtype | None, optional): The dtype of the model. Defaults to None.
        path_in_repo (str | None, optional): The path to the model in the repository. Defaults to None.
    Returns:
        EmbeddingModel: An initialized EmbeddingModel object.

    Atleast one of the following arguments must be provided:
        - model_name
        - hf_model_id

    If hf_model_id is provided, dtype is ignored and the path_in_repo has to be provided pointing to the model file in the repository.
    If model_name is provided, dtype is used to determine the model file to load.

    Example:
    ```python
    model = EmbeddingModel.from_pretrained_onnx(
        model=WhichModel.Bert,
        model_name=ONNXModel.BGESmallENV15Q,
        dtype=Dtype.Q4F16
    )

    model = EmbeddingModel.from_pretrained_onnx(
        model=WhichModel.Bert,
        hf_model_id="jinaai/jina-embeddings-v3",
        path_in_repo="onnx/model_fp16.onnx"
    )
    ```

    Note:
    This method loads a pre-trained model in ONNX format, which can offer improved inference speed
    compared to standard PyTorch models. ONNX models are particularly useful for deployment
    scenarios where performance is critical.
    """

ImageEmbedConfig

Represents the configuration for the Image Embedding model.

Attributes:

Name Type Description
buffer_size int | None

The buffer size for the Image Embedding model. Default is 100.
batch_size int | None

The batch size for processing the embeddings. Default is 32. Based on the memory, you can increase or decrease the batch size.
Source code in python/python/embed_anything/_embed_anything.pyi 
608
609
610
611
612
613
614
615
616
617
618
619
620
621
class ImageEmbedConfig:
    """
    Represents the configuration for the Image Embedding model.

    Attributes:
        buffer_size: The buffer size for the Image Embedding model. Default is 100.
        batch_size: The batch size for processing the embeddings. Default is 32. Based on the memory, you can increase or decrease the batch size.
    """

    def __init__(self, buffer_size: int | None = None, batch_size: int | None = None):
        self.buffer_size = buffer_size
        self.batch_size = batch_size
    buffer_size: int | None
    batch_size: int | None

ONNXModel

Bases: Enum

Enum representing various ONNX models.

| Enum Variant                     | Description                                      |
|----------------------------------|--------------------------------------------------|
| `AllMiniLML6V2`                  | sentence-transformers/all-MiniLM-L6-v2           |
| `AllMiniLML6V2Q`                 | Quantized sentence-transformers/all-MiniLM-L6-v2 |
| `AllMiniLML12V2`                 | sentence-transformers/all-MiniLM-L12-v2          |
| `AllMiniLML12V2Q`                | Quantized sentence-transformers/all-MiniLM-L12-v2|
| `ModernBERTBase`                 | nomic-ai/modernbert-embed-base                   |
| `ModernBERTLarge`                | nomic-ai/modernbert-embed-large                  |
| `BGEBaseENV15`                   | BAAI/bge-base-en-v1.5                            |
| `BGEBaseENV15Q`                  | Quantized BAAI/bge-base-en-v1.5                  |
| `BGELargeENV15`                  | BAAI/bge-large-en-v1.5                           |
| `BGELargeENV15Q`                 | Quantized BAAI/bge-large-en-v1.5                 |
| `BGESmallENV15`                  | BAAI/bge-small-en-v1.5 - Default                 |
| `BGESmallENV15Q`                 | Quantized BAAI/bge-small-en-v1.5                 |
| `NomicEmbedTextV1`               | nomic-ai/nomic-embed-text-v1                     |
| `NomicEmbedTextV15`              | nomic-ai/nomic-embed-text-v1.5                   |
| `NomicEmbedTextV15Q`             | Quantized nomic-ai/nomic-embed-text-v1.5         |
| `ParaphraseMLMiniLML12V2`        | sentence-transformers/paraphrase-MiniLM-L6-v2    |
| `ParaphraseMLMiniLML12V2Q`       | Quantized sentence-transformers/paraphrase-MiniLM-L6-v2 |
| `ParaphraseMLMpnetBaseV2`        | sentence-transformers/paraphrase-mpnet-base-v2   |
| `BGESmallZHV15`                  | BAAI/bge-small-zh-v1.5                           |
| `MultilingualE5Small`            | intfloat/multilingual-e5-small                   |
| `MultilingualE5Base`             | intfloat/multilingual-e5-base                    |
| `MultilingualE5Large`            | intfloat/multilingual-e5-large                   |
| `MxbaiEmbedLargeV1`              | mixedbread-ai/mxbai-embed-large-v1               |
| `MxbaiEmbedLargeV1Q`             | Quantized mixedbread-ai/mxbai-embed-large-v1     |
| `GTEBaseENV15`                   | Alibaba-NLP/gte-base-en-v1.5                     |
| `GTEBaseENV15Q`                  | Quantized Alibaba-NLP/gte-base-en-v1.5           |
| `GTELargeENV15`                  | Alibaba-NLP/gte-large-en-v1.5                    |
| `GTELargeENV15Q`                 | Quantized Alibaba-NLP/gte-large-en-v1.5          |
| `JINAV2SMALLEN`                  | jinaai/jina-embeddings-v2-small-en               |
| `JINAV2BASEEN`                   | jinaai/jina-embeddings-v2-base-en                |
| `JINAV3`                         | jinaai/jina-embeddings-v3                        |
| `SPLADEPPENV1`                   | prithivida/Splade_PP_en_v1                      |
| `SPLADEPPENV2`                   | prithivida/Splade_PP_en_v2                      |
Source code in python/python/embed_anything/_embed_anything.pyi 
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
class ONNXModel(Enum):
    """
    Enum representing various ONNX models.

    ```markdown
    | Enum Variant                     | Description                                      |
    |----------------------------------|--------------------------------------------------|
    | `AllMiniLML6V2`                  | sentence-transformers/all-MiniLM-L6-v2           |
    | `AllMiniLML6V2Q`                 | Quantized sentence-transformers/all-MiniLM-L6-v2 |
    | `AllMiniLML12V2`                 | sentence-transformers/all-MiniLM-L12-v2          |
    | `AllMiniLML12V2Q`                | Quantized sentence-transformers/all-MiniLM-L12-v2|
    | `ModernBERTBase`                 | nomic-ai/modernbert-embed-base                   |
    | `ModernBERTLarge`                | nomic-ai/modernbert-embed-large                  |
    | `BGEBaseENV15`                   | BAAI/bge-base-en-v1.5                            |
    | `BGEBaseENV15Q`                  | Quantized BAAI/bge-base-en-v1.5                  |
    | `BGELargeENV15`                  | BAAI/bge-large-en-v1.5                           |
    | `BGELargeENV15Q`                 | Quantized BAAI/bge-large-en-v1.5                 |
    | `BGESmallENV15`                  | BAAI/bge-small-en-v1.5 - Default                 |
    | `BGESmallENV15Q`                 | Quantized BAAI/bge-small-en-v1.5                 |
    | `NomicEmbedTextV1`               | nomic-ai/nomic-embed-text-v1                     |
    | `NomicEmbedTextV15`              | nomic-ai/nomic-embed-text-v1.5                   |
    | `NomicEmbedTextV15Q`             | Quantized nomic-ai/nomic-embed-text-v1.5         |
    | `ParaphraseMLMiniLML12V2`        | sentence-transformers/paraphrase-MiniLM-L6-v2    |
    | `ParaphraseMLMiniLML12V2Q`       | Quantized sentence-transformers/paraphrase-MiniLM-L6-v2 |
    | `ParaphraseMLMpnetBaseV2`        | sentence-transformers/paraphrase-mpnet-base-v2   |
    | `BGESmallZHV15`                  | BAAI/bge-small-zh-v1.5                           |
    | `MultilingualE5Small`            | intfloat/multilingual-e5-small                   |
    | `MultilingualE5Base`             | intfloat/multilingual-e5-base                    |
    | `MultilingualE5Large`            | intfloat/multilingual-e5-large                   |
    | `MxbaiEmbedLargeV1`              | mixedbread-ai/mxbai-embed-large-v1               |
    | `MxbaiEmbedLargeV1Q`             | Quantized mixedbread-ai/mxbai-embed-large-v1     |
    | `GTEBaseENV15`                   | Alibaba-NLP/gte-base-en-v1.5                     |
    | `GTEBaseENV15Q`                  | Quantized Alibaba-NLP/gte-base-en-v1.5           |
    | `GTELargeENV15`                  | Alibaba-NLP/gte-large-en-v1.5                    |
    | `GTELargeENV15Q`                 | Quantized Alibaba-NLP/gte-large-en-v1.5          |
    | `JINAV2SMALLEN`                  | jinaai/jina-embeddings-v2-small-en               |
    | `JINAV2BASEEN`                   | jinaai/jina-embeddings-v2-base-en                |
    | `JINAV3`                         | jinaai/jina-embeddings-v3                        |
    | `SPLADEPPENV1`                   | prithivida/Splade_PP_en_v1                      |
    | `SPLADEPPENV2`                   | prithivida/Splade_PP_en_v2                      |
    ```
    """

    AllMiniLML6V2 = "AllMiniLML6V2"

    AllMiniLML6V2Q = "AllMiniLML6V2Q"

    AllMiniLML12V2 = "AllMiniLML12V2"

    AllMiniLML12V2Q = "AllMiniLML12V2Q"

    ModernBERTBase = "ModernBERTBase"

    ModernBERTLarge = "ModernBERTLarge"

    BGEBaseENV15 = "BGEBaseENV15"

    BGEBaseENV15Q = "BGEBaseENV15Q"

    BGELargeENV15 = "BGELargeENV15"

    BGELargeENV15Q = "BGELargeENV15Q"

    BGESmallENV15 = "BGESmallENV15"

    BGESmallENV15Q = "BGESmallENV15Q"

    NomicEmbedTextV1 = "NomicEmbedTextV1"

    NomicEmbedTextV15 = "NomicEmbedTextV15"

    NomicEmbedTextV15Q = "NomicEmbedTextV15Q"

    ParaphraseMLMiniLML12V2 = "ParaphraseMLMiniLML12V2"

    ParaphraseMLMiniLML12V2Q = "ParaphraseMLMiniLML12V2Q"

    ParaphraseMLMpnetBaseV2 = "ParaphraseMLMpnetBaseV2"

    BGESmallZHV15 = "BGESmallZHV15"

    MultilingualE5Small = "MultilingualE5Small"

    MultilingualE5Base = "MultilingualE5Base"

    MultilingualE5Large = "MultilingualE5Large"

    MxbaiEmbedLargeV1 = "MxbaiEmbedLargeV1"

    MxbaiEmbedLargeV1Q = "MxbaiEmbedLargeV1Q"

    GTEBaseENV15 = "GTEBaseENV15"

    GTEBaseENV15Q = "GTEBaseENV15Q"

    GTELargeENV15 = "GTELargeENV15"

    GTELargeENV15Q = "GTELargeENV15Q"

    JINAV2SMALLEN = "JINAV2SMALLEN"

    JINAV2BASEEN = "JINAV2BASEEN"

    JINAV3 = "JINAV3"

    SPLADEPPENV1 = "SPLADEPPENV1"

    SPLADEPPENV2 = "SPLADEPPENV2"

Pooling

Bases: Enum

Represents the pooling method to use for the embedding model.

Source code in python/python/embed_anything/_embed_anything.pyi 
962
963
964
965
966
967
968
class Pooling(Enum):
    """
    Represents the pooling method to use for the embedding model.
    """
    Mean = "Mean"
    Cls = "Cls"    
    LastToken = "LastToken"

Reranker

Represents the Reranker model.

Source code in python/python/embed_anything/_embed_anything.pyi 
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
class Reranker:
    """
    Represents the Reranker model.
    """

    def __init__(
        self, model_id: str, revision: str | None = None, dtype: Dtype | None = None, path_in_repo: str | None = None
    ):
        """
        Initializes the Reranker object.
        """

    def from_pretrained(
        model_id: str, revision: str | None = None, dtype: Dtype | None = None, path_in_repo: str | None = None
    ) -> Reranker:
        """
        Loads a pre-trained Reranker model from the Hugging Face model hub.

        Args:
            model_id: The ID of the model from Hugging Face.
            revision: The revision of the model.
            dtype: The dtype of the model.
            path_in_repo: The path to the model in the repository.

        """

    def rerank(
        self, query: list[str], documents: list[str], batch_size: int
    ) -> RerankerResult:
        """
        Reranks the given documents for the query and returns a list of RerankerResult objects.

        Args:
            query: The query to rerank.
            documents: The list of documents to rerank.
            batch_size: The number of documents to process per batch.

        Returns:
            A list of RerankerResult objects.
        """

    def compute_scores(
        self, query: list[str], documents: list[str], batch_size: int
    ) -> list[list[float]]:
        """
        Computes the scores for the given query and documents.

        Args:
            query: The query to compute the scores for.
            documents: The list of documents to compute the scores for.
            batch_size: The batch size for processing the scores.

        Returns:
            A list of scores for the given query and documents.
        """

__init__(model_id, revision=None, dtype=None, path_in_repo=None)

Initializes the Reranker object.

Source code in python/python/embed_anything/_embed_anything.pyi 
468
469
470
471
472
473
def __init__(
    self, model_id: str, revision: str | None = None, dtype: Dtype | None = None, path_in_repo: str | None = None
):
    """
    Initializes the Reranker object.
    """

compute_scores(query, documents, batch_size)

Computes the scores for the given query and documents.

Parameters:

Name Type Description Default
query list[str]

The query to compute the scores for.

 required
documents list[str]

The list of documents to compute the scores for.

 required
batch_size int

The batch size for processing the scores.

 required

Returns:

Type Description
list[list[float]]

A list of scores for the given query and documents.
Source code in python/python/embed_anything/_embed_anything.pyi 
504
505
506
507
508
509
510
511
512
513
514
515
516
517
def compute_scores(
    self, query: list[str], documents: list[str], batch_size: int
) -> list[list[float]]:
    """
    Computes the scores for the given query and documents.

    Args:
        query: The query to compute the scores for.
        documents: The list of documents to compute the scores for.
        batch_size: The batch size for processing the scores.

    Returns:
        A list of scores for the given query and documents.
    """

from_pretrained(model_id, revision=None, dtype=None, path_in_repo=None)

Loads a pre-trained Reranker model from the Hugging Face model hub.

Parameters:

Name Type Description Default
model_id str

The ID of the model from Hugging Face.

 required
revision str | None

The revision of the model.

 None
dtype Dtype | None

The dtype of the model.

 None
path_in_repo str | None

The path to the model in the repository.

 None
Source code in python/python/embed_anything/_embed_anything.pyi 
475
476
477
478
479
480
481
482
483
484
485
486
487
def from_pretrained(
    model_id: str, revision: str | None = None, dtype: Dtype | None = None, path_in_repo: str | None = None
) -> Reranker:
    """
    Loads a pre-trained Reranker model from the Hugging Face model hub.

    Args:
        model_id: The ID of the model from Hugging Face.
        revision: The revision of the model.
        dtype: The dtype of the model.
        path_in_repo: The path to the model in the repository.

    """

rerank(query, documents, batch_size)

Reranks the given documents for the query and returns a list of RerankerResult objects.

Parameters:

Name Type Description Default
query list[str]

The query to rerank.

 required
documents list[str]

The list of documents to rerank.

 required
batch_size int

The number of documents to process per batch.

 required

Returns:

Type Description
RerankerResult

A list of RerankerResult objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
489
490
491
492
493
494
495
496
497
498
499
500
501
502
def rerank(
    self, query: list[str], documents: list[str], batch_size: int
) -> RerankerResult:
    """
    Reranks the given documents for the query and returns a list of RerankerResult objects.

    Args:
        query: The query to rerank.
        documents: The list of documents to rerank.
        batch_size: The number of documents to process per batch.

    Returns:
        A list of RerankerResult objects.
    """

RerankerResult

Represents the result of the reranking process.

Attributes:

Name Type Description
query str

The query to rerank.
documents list[DocumentRank]

The list of documents to rerank.
Source code in python/python/embed_anything/_embed_anything.pyi 
532
533
534
535
536
537
538
539
540
541
542
class RerankerResult:
    """
    Represents the result of the reranking process.

    Attributes:
        query: The query to rerank.
        documents: The list of documents to rerank.
    """

    query: str
    documents: list[DocumentRank]

TextEmbedConfig

Represents the configuration for the Text Embedding model.

Attributes:

Name Type Description
chunk_size int | None

The chunk size for the Text Embedding model. Default is 1000 Characters.
batch_size int | None

The batch size for processing the embeddings. Default is 32. Based on the memory, you can increase or decrease the batch size.
buffer_size int | None

The buffer size for the Text Embedding model. Default is 100.
late_chunking bool | None

A flag indicating whether to use late chunking for the Text Embedding model. Use late chunking to increase the context that is taken into account for each chunk. Default is False.
splitting_strategy str | None

The strategy to use for splitting the text into chunks. Default is "sentence". If semantic splitting is used, semantic_encoder is required.
semantic_encoder EmbeddingModel | None

The semantic encoder for the Text Embedding model. Default is None.
use_ocr bool | None

A flag indicating whether to use OCR for the Text Embedding model. Default is False.
tesseract_path str | None

The path to the Tesseract OCR executable. Default is None and uses the system path.
pdf_backend str | None

The backend to use for PDF text extraction. Currently only lopdf is supported. Default is lopdf.
Source code in python/python/embed_anything/_embed_anything.pyi 
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
class TextEmbedConfig:
    """
    Represents the configuration for the Text Embedding model.

    Attributes:
        chunk_size: The chunk size for the Text Embedding model. Default is 1000 Characters.
        batch_size: The batch size for processing the embeddings. Default is 32. Based on the memory, you can increase or decrease the batch size.
        buffer_size: The buffer size for the Text Embedding model. Default is 100.
        late_chunking: A flag indicating whether to use late chunking for the Text Embedding model. Use late chunking to increase the context that is taken into account for each chunk.  Default is False.
        splitting_strategy: The strategy to use for splitting the text into chunks. Default is "sentence". If semantic splitting is used, semantic_encoder is required.
        semantic_encoder: The semantic encoder for the Text Embedding model. Default is None.
        use_ocr: A flag indicating whether to use OCR for the Text Embedding model. Default is False.
        tesseract_path: The path to the Tesseract OCR executable. Default is None and uses the system path.
        pdf_backend: The backend to use for PDF text extraction. Currently only `lopdf` is supported. Default is `lopdf`.
    """

    def __init__(
        self,
        chunk_size: int | None = 1000,
        overlap_ratio: float | None = 0.0,
        batch_size: int | None = 32,
        late_chunking: bool | None = False,
        buffer_size: int | None = 100,
        splitting_strategy: str | None = "sentence",
        semantic_encoder: EmbeddingModel | None = None,
        use_ocr: bool | None = False,
        tesseract_path: str | None = None,
        pdf_backend: str | None = "lopdf",
    ):
        self.chunk_size = chunk_size
        self.overlap_ratio = overlap_ratio
        self.batch_size = batch_size
        self.late_chunking = late_chunking
        self.buffer_size = buffer_size
        self.splitting_strategy = splitting_strategy
        self.semantic_encoder = semantic_encoder
        self.use_ocr = use_ocr
        self.tesseract_path = tesseract_path
        self.pdf_backend = pdf_backend
    chunk_size: int | None
    overlap_ratio: float | None
    batch_size: int | None
    late_chunking: bool | None
    buffer_size: int | None
    splitting_strategy: str | None
    semantic_encoder: EmbeddingModel | None
    use_ocr: bool | None
    tesseract_path: str | None
    pdf_backend: str | None

VideoEmbedConfig

Represents the configuration for the Video Embedding model.

Attributes:

Name Type Description
frame_step int | None

Sample every Nth frame. Default is 30.
max_frames int | None

Maximum number of frames to embed. Default is 300.
batch_size int | None

The batch size for processing frames. Default is 32.
Source code in python/python/embed_anything/_embed_anything.pyi 
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
class VideoEmbedConfig:
    """
    Represents the configuration for the Video Embedding model.

    Attributes:
        frame_step: Sample every Nth frame. Default is 30.
        max_frames: Maximum number of frames to embed. Default is 300.
        batch_size: The batch size for processing frames. Default is 32.
    """

    def __init__(
        self,
        frame_step: int | None = None,
        max_frames: int | None = None,
        batch_size: int | None = None,
    ):
        self.frame_step = frame_step
        self.max_frames = max_frames
        self.batch_size = batch_size
    frame_step: int | None
    max_frames: int | None
    batch_size: int | None

embed_audio_file(file_path, audio_decoder, embedder, text_embed_config=TextEmbedConfig(chunk_size=1000, batch_size=32))

Embeds the given audio file and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the audio file to embed.

 required
audio_decoder AudioDecoderModel

The audio decoder model to use.

 required
embedder EmbeddingModel

The embedding model to use.

 required
text_embed_config TextEmbedConfig | None

The configuration for the embedding model.

 TextEmbedConfig(chunk_size=1000, batch_size=32)

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.

Example: 

import embed_anything
audio_decoder = embed_anything.AudioDecoderModel.from_pretrained_hf(
    "openai/whisper-tiny.en", revision="main", model_type="tiny-en", quantized=False
)

embedder = embed_anything.EmbeddingModel.from_pretrained_hf(
    embed_anything.WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)

config = embed_anything.TextEmbedConfig(chunk_size=1000, batch_size=32)
data = embed_anything.embed_audio_file(
    "test_files/audio/samples_hp0.wav",
    audio_decoder=audio_decoder,
    embedder=embedder,
    text_embed_config=config,
)

Source code in python/python/embed_anything/_embed_anything.pyi 
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
def embed_audio_file(
    file_path: str,
    audio_decoder: AudioDecoderModel,
    embedder: EmbeddingModel,
    text_embed_config: TextEmbedConfig | None = TextEmbedConfig(
        chunk_size=1000, batch_size=32
    ),
) -> list[EmbedData]:
    """
    Embeds the given audio file and returns a list of EmbedData objects.

    Args:
        file_path: The path to the audio file to embed.
        audio_decoder: The audio decoder model to use.
        embedder: The embedding model to use.
        text_embed_config: The configuration for the embedding model.

    Returns:
        A list of EmbedData objects.

    Example:
    ```python

    import embed_anything
    audio_decoder = embed_anything.AudioDecoderModel.from_pretrained_hf(
        "openai/whisper-tiny.en", revision="main", model_type="tiny-en", quantized=False
    )

    embedder = embed_anything.EmbeddingModel.from_pretrained_hf(
        embed_anything.WhichModel.Bert,
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
    )

    config = embed_anything.TextEmbedConfig(chunk_size=1000, batch_size=32)
    data = embed_anything.embed_audio_file(
        "test_files/audio/samples_hp0.wav",
        audio_decoder=audio_decoder,
        embedder=embedder,
        text_embed_config=config,
    )
    ```
    """

embed_directory(file_path, embedder, extensions, config=None, adapter=None)

Embeds the files in the given directory and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the directory containing the files to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
extensions list[str]

The list of file extensions to consider for embedding.

 required
config TextEmbedConfig | None

The configuration for the embedding model.

 None
adapter Adapter | None

The adapter to use for storing the embeddings in a vector database.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.

Example: 

import embed_anything
model = embed_anything.EmbeddingModel.from_pretrained_hf(
    embed_anything.WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)
data = embed_anything.embed_directory("test_files", embedder=model, extensions=[".pdf"])

Source code in python/python/embed_anything/_embed_anything.pyi 
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
def embed_directory(
    file_path: str,
    embedder: EmbeddingModel,
    extensions: list[str],
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the files in the given directory and returns a list of EmbedData objects.

    Args:
        file_path: The path to the directory containing the files to embed.
        embedder: The embedding model to use.
        extensions: The list of file extensions to consider for embedding.
        config: The configuration for the embedding model.
        adapter: The adapter to use for storing the embeddings in a vector database.

    Returns:
        A list of EmbedData objects.

    Example:
    ```python
    import embed_anything
    model = embed_anything.EmbeddingModel.from_pretrained_hf(
        embed_anything.WhichModel.Bert,
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
    )
    data = embed_anything.embed_directory("test_files", embedder=model, extensions=[".pdf"])
    ```
    """

embed_file(file_path, embedder, config=None, adapter=None)

Embeds the given file and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the file to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
config TextEmbedConfig | None

The configuration for the embedding model.

 None
adapter Adapter | None

The adapter to use for storing the embeddings in a vector database.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.

Example: 

import embed_anything
model = embed_anything.EmbeddingModel.from_pretrained_hf(
    embed_anything.WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)
data = embed_anything.embed_file("test_files/test.pdf", embedder=model)

Source code in python/python/embed_anything/_embed_anything.pyi 
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
def embed_file(
    file_path: str,
    embedder: EmbeddingModel,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given file and returns a list of EmbedData objects.

    Args:
        file_path: The path to the file to embed.
        embedder: The embedding model to use.
        config: The configuration for the embedding model.
        adapter: The adapter to use for storing the embeddings in a vector database.

    Returns:
        A list of EmbedData objects.

    Example:
    ```python
    import embed_anything
    model = embed_anything.EmbeddingModel.from_pretrained_hf(
        embed_anything.WhichModel.Bert,
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
    )
    data = embed_anything.embed_file("test_files/test.pdf", embedder=model)
    ```
    """

embed_files_batch(files, embedder, config=None, adapter=None)

Embeds the given files and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
files list[str]

The list of files to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
config TextEmbedConfig | None

The configuration for the embedding model.

 None
adapter Adapter | None

The adapter to use for storing the embeddings in a vector database.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.

Example: 

import embed_anything
model = embed_anything.EmbeddingModel.from_pretrained_hf(
    embed_anything.WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)
data = embed_anything.embed_files_batch(
    ["test_files/test.pdf", "test_files/test.txt"],
    embedder=model,
    config=embed_anything.TextEmbedConfig(),
    adapter=None,
)

Source code in python/python/embed_anything/_embed_anything.pyi 
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
def embed_files_batch(
    files: list[str],
    embedder: EmbeddingModel,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given files and returns a list of EmbedData objects.

    Args:
        files: The list of files to embed.
        embedder: The embedding model to use.
        config: The configuration for the embedding model.
        adapter: The adapter to use for storing the embeddings in a vector database.

    Returns:
        A list of EmbedData objects.

    Example:
    ```python
    import embed_anything
    model = embed_anything.EmbeddingModel.from_pretrained_hf(
        embed_anything.WhichModel.Bert,
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
    )
    data = embed_anything.embed_files_batch(
        ["test_files/test.pdf", "test_files/test.txt"],
        embedder=model,
        config=embed_anything.TextEmbedConfig(),
        adapter=None,
    )
    ```
    """

embed_html(file_name, embedder, origin=None, config=None, adapter=None)

Embeds the given HTML file and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_name str

The path to the HTML file to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
origin str | None

The origin of the HTML file.

 None
config TextEmbedConfig | None

The configuration for the embedding model.

 None
adapter Adapter | None

The adapter to use for storing the embeddings.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.

Example: 

import embed_anything
model = embed_anything.EmbeddingModel.from_pretrained_hf(
    embed_anything.WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)
data = embed_anything.embed_html(
    "test_files/test.html", embedder=model, origin="https://www.akshaymakes.com/"
)

Source code in python/python/embed_anything/_embed_anything.pyi 
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
def embed_html(
    file_name: str,
    embedder: EmbeddingModel,
    origin: str | None = None,
    config: TextEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the given HTML file and returns a list of EmbedData objects.

    Args:
        file_name: The path to the HTML file to embed.
        embedder: The embedding model to use.
        origin: The origin of the HTML file.
        config: The configuration for the embedding model.
        adapter: The adapter to use for storing the embeddings.

    Returns:
        A list of EmbedData objects.

    Example:
    ```python
    import embed_anything
    model = embed_anything.EmbeddingModel.from_pretrained_hf(
        embed_anything.WhichModel.Bert,
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
    )
    data = embed_anything.embed_html(
        "test_files/test.html", embedder=model, origin="https://www.akshaymakes.com/"
    )
    ```
    """

embed_image_directory(file_path, embedder, config=None, adapter=None)

Embeds the images in the given directory and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the directory containing the images to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
config ImageEmbedConfig | None

The configuration for the embedding model.

 None
adapter Adapter | None

The adapter to use for storing the embeddings in a vector database.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
def embed_image_directory(
    file_path: str,
    embedder: EmbeddingModel,
    config: ImageEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData]:
    """
    Embeds the images in the given directory and returns a list of EmbedData objects.

    Args:
        file_path: The path to the directory containing the images to embed.
        embedder: The embedding model to use.
        config: The configuration for the embedding model.
        adapter: The adapter to use for storing the embeddings in a vector database.

    Returns:
        A list of EmbedData objects.
    """

embed_query(query, embedder, config=None)

Embeds the given query and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
query list[str]

The query to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
config TextEmbedConfig | None

The configuration for the embedding model.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.

Example:

import embed_anything
model = embed_anything.EmbeddingModel.from_pretrained_hf(
    embed_anything.WhichModel.Bert,
    model_id="sentence-transformers/all-MiniLM-L6-v2",
    revision="main",
)
Source code in python/python/embed_anything/_embed_anything.pyi 
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
def embed_query(
    query: list[str], embedder: EmbeddingModel, config: TextEmbedConfig | None = None
) -> list[EmbedData]:
    """
    Embeds the given query and returns a list of EmbedData objects.

    Args:
        query: The query to embed.
        embedder: The embedding model to use.
        config: The configuration for the embedding model.

    Returns:
        A list of EmbedData objects.

    Example:

    ```python
    import embed_anything
    model = embed_anything.EmbeddingModel.from_pretrained_hf(
        embed_anything.WhichModel.Bert,
        model_id="sentence-transformers/all-MiniLM-L6-v2",
        revision="main",
    )
    ```
    """

embed_video_directory(file_path, embedder, config=None, adapter=None)

Embeds all videos in the given directory and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the directory containing videos to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
config VideoEmbedConfig | None

The configuration for video embedding.

 None
adapter Adapter | None

The adapter to use for storing the embeddings in a vector database.

 None

Returns:

Type Description
list[EmbedData] | None

A list of EmbedData objects, or None if an adapter is used.
Source code in python/python/embed_anything/_embed_anything.pyi 
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
def embed_video_directory(
    file_path: str,
    embedder: EmbeddingModel,
    config: VideoEmbedConfig | None = None,
    adapter: Adapter | None = None,
) -> list[EmbedData] | None:
    """
    Embeds all videos in the given directory and returns a list of EmbedData objects.

    Args:
        file_path: The path to the directory containing videos to embed.
        embedder: The embedding model to use.
        config: The configuration for video embedding.
        adapter: The adapter to use for storing the embeddings in a vector database.

    Returns:
        A list of EmbedData objects, or None if an adapter is used.
    """

embed_video_file(file_path, embedder, config=None)

Embeds the given video file by sampling frames and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
file_path str

The path to the video file to embed.

 required
embedder EmbeddingModel

The embedding model to use.

 required
config VideoEmbedConfig | None

The configuration for video embedding.

 None

Returns:

Type Description
list[EmbedData]

A list of EmbedData objects.
Source code in python/python/embed_anything/_embed_anything.pyi 
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
def embed_video_file(
    file_path: str,
    embedder: EmbeddingModel,
    config: VideoEmbedConfig | None = None,
) -> list[EmbedData]:
    """
    Embeds the given video file by sampling frames and returns a list of EmbedData objects.

    Args:
        file_path: The path to the video file to embed.
        embedder: The embedding model to use.
        config: The configuration for video embedding.

    Returns:
        A list of EmbedData objects.
    """

embed_webpage(url, embedder, config, adapter)

Embeds the webpage at the given URL and returns a list of EmbedData objects.

Parameters:

Name Type Description Default
url str

The URL of the webpage to embed.

 required
embedder EmbeddingModel

The name of the embedding model to use. Choose between "OpenAI", "Jina", "Bert"

 required
config TextEmbedConfig | None

The configuration for the embedding model.

 required
adapter Adapter | None

The adapter to use for storing the embeddings.

 required

Returns:

Type Description
list[EmbedData] | None

A list of EmbedData objects

Example: 

import embed_anything

config = embed_anything.EmbedConfig(
    openai_config=embed_anything.OpenAIConfig(model="text-embedding-3-small")
)
data = embed_anything.embed_webpage(
    "https://www.akshaymakes.com/", embedder="OpenAI", config=config
)

Source code in python/python/embed_anything/_embed_anything.pyi 
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
def embed_webpage(
    url: str,
    embedder: EmbeddingModel,
    config: TextEmbedConfig | None,
    adapter: Adapter | None,
) -> list[EmbedData] | None:
    """Embeds the webpage at the given URL and returns a list of EmbedData
    objects.

    Args:
        url: The URL of the webpage to embed.
        embedder: The name of the embedding model to use. Choose between "OpenAI", "Jina", "Bert"
        config: The configuration for the embedding model.
        adapter: The adapter to use for storing the embeddings.

    Returns:
        A list of EmbedData objects

    Example:
    ```python
    import embed_anything

    config = embed_anything.EmbedConfig(
        openai_config=embed_anything.OpenAIConfig(model="text-embedding-3-small")
    )
    data = embed_anything.embed_webpage(
        "https://www.akshaymakes.com/", embedder="OpenAI", config=config
    )
    ```
    """