Skip to content

Openai embedding driver

OpenAiEmbeddingDriver

Bases: BaseEmbeddingDriver

Attributes:

Name Type Description
model str

OpenAI embedding model name. Defaults to text-embedding-3-small.
base_url Optional[str]

API URL. Defaults to OpenAI's v1 API URL.
api_key Optional[str]

API key to pass directly. Defaults to OPENAI_API_KEY environment variable.
organization Optional[str]

OpenAI organization. Defaults to 'OPENAI_ORGANIZATION' environment variable.
tokenizer OpenAiTokenizer

Optionally provide custom OpenAiTokenizer.
client OpenAI

Optionally provide custom openai.OpenAI client.
azure_deployment OpenAI

An Azure OpenAi deployment id.
azure_endpoint OpenAI

An Azure OpenAi endpoint.
azure_ad_token OpenAI

An optional Azure Active Directory token.
azure_ad_token_provider OpenAI

An optional Azure Active Directory token provider.
api_version OpenAI

An Azure OpenAi API version.
Source code in griptape/drivers/embedding/openai_embedding_driver.py 
 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
@define
class OpenAiEmbeddingDriver(BaseEmbeddingDriver):
    """
    Attributes:
        model: OpenAI embedding model name. Defaults to `text-embedding-3-small`.
        base_url: API URL. Defaults to OpenAI's v1 API URL.
        api_key: API key to pass directly. Defaults to `OPENAI_API_KEY` environment variable.
        organization: OpenAI organization. Defaults to 'OPENAI_ORGANIZATION' environment variable.
        tokenizer: Optionally provide custom `OpenAiTokenizer`.
        client: Optionally provide custom `openai.OpenAI` client.
        azure_deployment: An Azure OpenAi deployment id.
        azure_endpoint: An Azure OpenAi endpoint.
        azure_ad_token: An optional Azure Active Directory token.
        azure_ad_token_provider: An optional Azure Active Directory token provider.
        api_version: An Azure OpenAi API version.
    """

    DEFAULT_MODEL = "text-embedding-3-small"

    model: str = field(default=DEFAULT_MODEL, kw_only=True, metadata={"serializable": True})
    base_url: Optional[str] = field(default=None, kw_only=True, metadata={"serializable": True})
    api_key: Optional[str] = field(default=None, kw_only=True, metadata={"serializable": False})
    organization: Optional[str] = field(default=None, kw_only=True, metadata={"serializable": True})
    client: openai.OpenAI = field(
        default=Factory(
            lambda self: openai.OpenAI(api_key=self.api_key, base_url=self.base_url, organization=self.organization),
            takes_self=True,
        )
    )
    tokenizer: OpenAiTokenizer = field(
        default=Factory(lambda self: OpenAiTokenizer(model=self.model), takes_self=True), kw_only=True
    )

    def try_embed_chunk(self, chunk: str) -> list[float]:
        # Address a performance issue in older ada models
        # https://github.com/openai/openai-python/issues/418#issuecomment-1525939500
        if self.model.endswith("001"):
            chunk = chunk.replace("\n", " ")
        return self.client.embeddings.create(**self._params(chunk)).data[0].embedding

    def _params(self, chunk: str) -> dict:
        return {"input": chunk, "model": self.model}

DEFAULT_MODEL = 'text-embedding-3-small' class-attribute instance-attribute

api_key: Optional[str] = field(default=None, kw_only=True, metadata={'serializable': False}) class-attribute instance-attribute

base_url: Optional[str] = field(default=None, kw_only=True, metadata={'serializable': True}) class-attribute instance-attribute

client: openai.OpenAI = field(default=Factory(lambda self: openai.OpenAI(api_key=self.api_key, base_url=self.base_url, organization=self.organization), takes_self=True)) class-attribute instance-attribute

model: str = field(default=DEFAULT_MODEL, kw_only=True, metadata={'serializable': True}) class-attribute instance-attribute

organization: Optional[str] = field(default=None, kw_only=True, metadata={'serializable': True}) class-attribute instance-attribute

tokenizer: OpenAiTokenizer = field(default=Factory(lambda self: OpenAiTokenizer(model=self.model), takes_self=True), kw_only=True) class-attribute instance-attribute

try_embed_chunk(chunk)

Source code in griptape/drivers/embedding/openai_embedding_driver.py 
42
43
44
45
46
47
def try_embed_chunk(self, chunk: str) -> list[float]:
    # Address a performance issue in older ada models
    # https://github.com/openai/openai-python/issues/418#issuecomment-1525939500
    if self.model.endswith("001"):
        chunk = chunk.replace("\n", " ")
    return self.client.embeddings.create(**self._params(chunk)).data[0].embedding