Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
api_key | Optional[str] | None | Azure OpenAI API key |
azure_endpoint | Optional[str] | None | Azure OpenAI endpoint URL |
api_version | str | "2024-02-01" | Azure OpenAI API version |
deployment_name | Optional[str] | None | Azure deployment name |
use_managed_identity | bool | False | Use Azure Managed Identity |
tenant_id | Optional[str] | None | Azure tenant ID |
client_id | Optional[str] | None | Azure client ID for managed identity |
model_name | str | "text-embedding-ada-002" | Embedding model name |
enable_content_filtering | bool | True | Enable Azure content filtering |
data_residency_region | Optional[str] | None | Data residency region |
parallel_requests | int | 3 | Parallel requests (Azure has lower limits) |
requests_per_minute | int | 240 | Requests per minute for Azure |
tokens_per_minute | int | 240000 | Tokens per minute for Azure |
Functions
__init__
Initialize the AzureOpenAIEmbedding provider.
Parameters:
config(Optional[AzureOpenAIEmbeddingConfig]): Configuration object**kwargs: Additional configuration options
_setup_managed_identity
Setup Azure Managed Identity authentication.
_setup_client
Setup the Azure OpenAI client with proper configuration.
_get_azure_token
Get Azure AD token for authentication.
Returns:
str: Azure AD token
supported_modes
Get supported embedding modes.
Returns:
List[EmbeddingMode]: List of supported embedding modes
pricing_info
Get Azure OpenAI embedding pricing (may differ from standard OpenAI).
Returns:
Dict[str, float]: Pricing information
get_model_info
Get information about the current Azure OpenAI model.
Returns:
Dict[str, Any]: Model information
_check_rate_limits
Check and enforce Azure-specific rate limits.
Parameters:
estimated_tokens(int): Estimated token count
_estimate_tokens
Estimate token count for texts.
Parameters:
texts(List[str]): List of texts
int: Estimated token count
_embed_batch
Embed a batch of texts using Azure OpenAI API.
Parameters:
texts(List[str]): List of text strings to embedmode(EmbeddingMode): Embedding mode (not used by Azure OpenAI, but kept for compatibility)
List[List[float]]: List of embedding vectors
validate_connection
Validate Azure OpenAI connection and deployment access.
Returns:
bool: True if connection is valid
get_azure_info
Get Azure-specific configuration information.
Returns:
Dict[str, Any]: Azure configuration information
get_compliance_info
Get compliance and security information.
Returns:
Dict[str, Any]: Compliance information
close
Clean up Azure OpenAI client connections and resources.
create_azure_openai_embedding
Quick factory function for Azure OpenAI embeddings.
Parameters:
azure_endpoint(str): Azure OpenAI endpoint URLdeployment_name(str): Azure deployment nameapi_key(Optional[str]): Azure OpenAI API key (optional if using managed identity)use_managed_identity(bool): Use Azure Managed Identity authentication**kwargs: Additional configuration options
AzureOpenAIEmbedding: Configured AzureOpenAIEmbedding instance
create_azure_embedding_with_managed_identity
Create Azure OpenAI embedding with managed identity.
Parameters:
azure_endpoint(str): Azure OpenAI endpoint URLdeployment_name(str): Azure deployment name**kwargs: Additional configuration options
AzureOpenAIEmbedding: Configured AzureOpenAIEmbedding instance