AzureOpenAIEmbeddings - Can't Avoid Rate Limit

[yuval-illumex]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

return AzureOpenAIEmbeddings(
            openai_api_key=azure_config["api_key"],
            openai_api_version=azure_config["api_version"],
            azure_endpoint=azure_config["api_endpoint"],
            deployment="text-embedding-ada-002",
            max_retries=5,
        )

Description

When trying to using AzureOpenAI for embeddings, i'm getting rate limit error while i'm maxed out of my quota.
I want to make the embeddings to wait for a minute, in order to pass the rate limit per minute.
I really wanted to use retry_min_seconds
however looks like it's only available for OpenAI.

Is retry_min_seconds available in AzureOpenAI in some way?
Is there another alternative (I didn't found one)?

Thanks!

System Info

langchain 0.1.16/langhain_openai 0.1.7
linux+mac
python 3.11

[yuval-illumex]

Adding updated info:

I ran this in Debug log level and i'm seeing this log:

DEBUG:openai._base_client:Sending HTTP Request: POST https://lmx-eastus2-prod.openai.azure.com//openai/deployments/text-embedding-ada-002/embeddings?api-version=2024-10-21
DEBUG:httpcore.http11:send_request_headers.started request=<Request [b'POST']>
DEBUG:httpcore.http11:send_request_headers.complete
DEBUG:httpcore.http11:send_request_body.started request=<Request [b'POST']>
DEBUG:httpcore.http11:send_request_body.complete
DEBUG:httpcore.http11:receive_response_headers.started request=<Request [b'POST']>
DEBUG:httpcore.http11:receive_response_headers.complete return_value=(b'HTTP/1.1', 429, b'Too Many Requests', [(b'Content-Length', b'346'), (b'Content-Type', b'application/json'), (b'Retry-After', b'86400'), (b'x-ratelimit-reset-tokens', b'86400'), (b'apim-request-id', b'7b1f95d4-e813-4319-97ec-48fa967de438'), (b'Strict-Transport-Security', b'max-age=31536000; includeSubDomains; preload'), (b'x-content-type-options', b'nosniff'), (b'policy-id', b'EmbedDeploymentRatelimit-Token'), (b'x-ms-region', b'East US 2'), (b'x-ratelimit-remaining-requests', b'327'), (b'Date', b'Mon, 27 Jan 2025 11:06:15 GMT')])

1. Looks like Azure OpenAI returned 'Retry-After' 24 hours, however if I run this after 1 minute it works, am I missing something?
2. AzureOpenAIEmbeddings ignoring the 'Retry-After' header (although openai module is supporting it?)

Please, any help will be appreciated as i'm stuck

[yuval-illumex]

EDIT: I had increase in data size which caused the data to be sent as too big.
Dividing it into smaller chunks fixed this.
Thanks.