Skip to content

Azure AI Content Understanding (ACU) — Intermittent Timeout Errors with GPT-4.1-mini Custom Analyzer Model #47942

Description

@Jivitesh2001
  • Package Name: azure-ai-contentunderstanding
  • Package Version: 1.1.0
  • Operating System: Using Docker-image - python:3.11-slim with k8s
  • Python Version: 3.11

Describe the bug
We are using Azure AI Content Understanding(ACU) to extract structured data from documents. Requests are submitted asynchronously via the Python SDK (begin_analyse), and results are retrieved through polling on the returned AsyncLROPoller. When multiple documents are submitted for processing at around the same time, a subset of the ACU requests intermittently time out.
Resubmitting a timed-out file in isolation succeeds, and the response is typically returned within 10–12 seconds. This indicates the issue is not related to the content of any specific file, but rather to how the service behaves under concurrent load.

Troubleshooting Performed -

We suspected the timeouts were caused by request concurrency (a "thundering herd" effect on the ACU service) and tried the following mitigations, none of which resolved the issue:

  1. Increased the SDK polling interval from 1s to 5s to reduce polling load.
  2. Increased the Tokens Per Minute (TPM) quota on the underlying gpt-4.1-mini deployment from 250K to 35M TPM.
  3. Reduced the number of concurrent requests sent to ACU from 10 to 5 (this reduced but did not eliminate the errors, and increased overall processing latency), with exponential back-off retries.
  4. Added random jitter to concurrent requests so they would not fire at the exact same time.
  5. Added a rate-limiting mechanism to space out request submissions to ACU, with small randomized delays (0–200ms) between each request.

Key Finding -

We changed the chat completion model used by the ACU analyzer from gpt-4.1-mini to gpt-5.2 and gpt-4.1, and the timeout errors stopped occurring — under the same concurrency and request volume that previously produced timeouts.
This strongly suggests the root cause is specific to the gpt-4.1-mini deployment (or how ACU analyzers utilize that model), rather than general request concurrency on our end.

Request to Azure Support -

Could you help us understand:

  1. Whether there is a known issue or capacity constraint with ACU analyzers backed by gpt-4.1-mini under concurrent load.
  2. Whether gpt-4.1-mini deployments used by ACU have different throughput/latency characteristics or throttling behavior compared to gpt-5.2 / gpt-4.1 that would explain this.
  3. Any recommended configuration (deployment SKU, region, capacity type) to make gpt-4.1-mini reliable for this workload?

Metadata

Metadata

Assignees

No one assigned

    Labels

    ClientThis issue points to a problem in the data-plane of the library.Cognitive - Content Understandingcustomer-reportedIssues that are reported by GitHub users external to the Azure organization.needs-team-triageWorkflow: This issue needs the team to triage.questionThe issue doesn't require a change to the product in order to be resolved. Most issues start as that

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions