16 views
-/https://github.com/berriai/litellm/issues/19334
GitHub Ā· issue

#19334 [Bug]: Calls to azure_ai FLUX-1.1-pro and FLUX.1-Kontext-pro fail after upgrade to 1.80.15

  • State: open
  • Author: @OQCDoe
  • Labels: bug, proxy, llm translation

### Check for existing issues

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

### What happened?

After upgrading our instance to 1.80.15-stable.1, all calls to our Azure AI Foundry deployments of FLUX-1.1-pro and FLUX.1-Kontext-pro are failing. Previous version was 1.80.11

Config: ```yaml - model_name: flux1-kontext-pro litellm_params: model: azure_ai/FLUX.1-Kontext-pro api_base: "https://company.services.ai.azure.com/" api_key: os.environ/TOKEN api_version: 2025-04-01-preview rpm: 30 additional_drop_params: ["response_format"] model_info: mode: image_generation base_model: azure_ai/FLUX.1-Kontext-pro - model_name: flux11-pro litellm_params: model: azure_ai/FLUX-1.1-pro api_base: "https://company.services.ai.azure.com/" api_key: os.environ/TOKEN api_version: 2025-04-01-preview rpm: 30 additional_drop_params: ["response_format"] model_info: mode: image_generation base_model: azure_ai/FLUX-1.1-pro - model_name: flux2-pro litellm_params: model: azure_ai/flux.2-pro api_base: "https://company.services.ai.azure.c…

GitHub resolver

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

Refresh page
vote history (1 events)
#0 of 0 Ā· 31d18h5m56s ago — entered Ā· #import:https:::github.com:berriai:litellm post #2752
The left issue is harder because it involves diagnosing a regression in a provider-specific image-generation integration across multiple model variants and API behaviors, with greater external-service and compatibility uncertainty. The right issue appears more localized to shared parameter-routing logic, with a clearer implementation path and narrower test surface.
discussed in #import:https:::github.com:berriai:litellm

ranked child groups

no voted pairs yet in this scope

cli
src
spread
search