Python: Azure OpenAI text embedding function does not cap inputs #2099
Description
Describe the bug
In python using the AzureTextEmbedding:generate_embeddings_async() method with more than 16 elements in the texts list will generate an error because the API doesn't support more than 16.
To Reproduce
Steps to reproduce the behavior:
import asyncio
from semantic_kernel.connectors.ai.open_ai import AzureTextEmbedding
async def embed_test(texts):
generator = AzureTextEmbedding(
deployment_name=AOAI_TEXT_EMBEDDING,
endpoint=AOAI_URL,
api_key=AOAI_KEY,
)
return await generator.generate_embeddings_async(texts)
asyncio.run(embed_test(texts=["hello"]*20))
generates:
semantic_kernel/connectors/ai/open_ai/services/open_ai_text_embedding.py", line 77, in generate_embeddings_async
raise AIException(
semantic_kernel.connectors.ai.ai_exception.AIException: (<ErrorCodes.ServiceError: 6>, 'OpenAI service failed to generate embeddings', InvalidRequestError(message='Too many inputs. The max number of inputs is 16. We hope to increase the number of inputs per request soon. Please contact us through an Azure support request at: https://go.microsoft.com/fwlink/?linkid= for further questions.', param=None, code=None, http_status=400, request_id=None))
Expected behavior
It would be nice to provide a way of processing more than 16 inputs
Platform
- OS: Ubuntu [WSL]
- IDE: VS Code
- Language: Python
- Source: main branch of repository [5721ac4]
Metadata
Metadata
Assignees
Labels
Type
Projects
Status