huggingface_endpoint.py — langchain Source File

Source Code

from __future__ import annotations

import os
from typing import Any

from langchain_core.embeddings import Embeddings
from langchain_core.utils import from_env
from pydantic import BaseModel, ConfigDict, Field, model_validator
from typing_extensions import Self

DEFAULT_MODEL = "sentence-transformers/all-mpnet-base-v2"
VALID_TASKS = ("feature-extraction",)


class HuggingFaceEndpointEmbeddings(BaseModel, Embeddings):
    """HuggingFaceHub embedding models.

    To use, you should have the `huggingface_hub` python package installed, and the
    environment variable `HUGGINGFACEHUB_API_TOKEN` set with your API token, or pass
    it as a named parameter to the constructor.

    Example:
        ```python
        from langchain_huggingface import HuggingFaceEndpointEmbeddings

        model = "sentence-transformers/all-mpnet-base-v2"
        hf = HuggingFaceEndpointEmbeddings(
            model=model,
            task="feature-extraction",
            huggingfacehub_api_token="my-api-key",
        )
        ```
    """

    client: Any = None

    async_client: Any = None

    model: str | None = None
    """Model name to use."""

    provider: str | None = None
    """Name of the provider to use for inference with the model specified in
        `repo_id`. e.g. "sambanova". if not specified, defaults to HF Inference API.
        available providers can be found in the [huggingface_hub documentation](https://huggingface.co/docs/huggingface_hub/guides/inference#supported-providers-and-tasks)."""

    repo_id: str | None = None
    """Huggingfacehub repository id, for backward compatibility."""

    task: str | None = "feature-extraction"
    """Task to call the model with."""

    model_kwargs: dict | None = None
    """Keyword arguments to pass to the model."""

    huggingfacehub_api_token: str | None = Field(
        default_factory=from_env("HUGGINGFACEHUB_API_TOKEN", default=None)
    )

    model_config = ConfigDict(
// ... (113 more lines)