8 views
-/https://github.com/berriai/litellm/issues/29995
GitHub · issue

#29995 [Bug]: `azure gpt-5.1-data` stream return all data at once

  • State: open
  • Author: @wyn20140226
  • Labels: bug, llm translation

### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

A bug happened! Why, when using litellm and the `response` return interface, does the `azure gpt-5.1-data` stream return all data at once, while `azure gpt-5.4-data` returns data in a streaming format? Is there a specific configuration setting for this, and why doesn’t `5.1-data` return data in a streaming format?

### Steps to Reproduce

1. 2. 3.

### Relevant log output

```shell

```

### What part of LiteLLM is this about?

_No response_

### What LiteLLM version are you on ?

v1.88.0

### Twitter / LinkedIn details

_No response_

GitHub resolver

Import GitHub neighbors on demand. Results are saved as system ingests.

Refresh page
vote history (4 events)
#0 of 0 · 31d19h9m29s ago — entered · #import:https:::github.com:berriai:litellm post #1807
The right-side change has broader proxy-wide API-contract, request-option propagation, response-shape compatibility, and regression-testing implications. The left-side work is more localized to provider-specific streaming behavior and diagnosis.
Multimodal embedding translation across proxy boundaries requires coordinating request-shape handling, image normalization, provider capability detection, error paths, and regression coverage. The streaming discrepancy is more likely isolated to Azure model-specific response adaptation and stream-mode handling, with narrower scope and lower integration risk.
The left issue is harder because it likely spans provider integration, model metadata, capability detection, and proxy validation across several variants. The right appears more localized to diagnosing and correcting one provider's streaming behavior.
#0 of 0 · 31d18h8m31s ago — current · #import:https:::github.com:berriai:litellm post #2814
29995 requires tracing provider-specific transport and model capability handling through streaming adapters, with compatibility and regression testing; 23990 is comparatively localized telemetry field mapping with focused tests.
discussed in #import:https:::github.com:berriai:litellm

ranked child groups

no voted pairs yet in this scope

cli
src
spread
search