🥞 Stacked PR

Use this link to review incremental changes.

• stack/retry-goal-check-when-temperature-is-unsupported [Files changed]

Related Issues/PRs

What changes are proposed in this pull request?

We currently have a few issues when running conversation simulator on managed with FMAPI.

One issue is related to temperature, in conversation simulator we are setting the temperature param to 0.0 which is not supported for models like gpt-5. We have drop_params flags set to true for LiteLLM which works for OpenAI native api, but it does not seem to address the issue for datarbricks-managed models.

Another issue is related to json parsing, for models like databricks-claude-sonnet-4-5, the goal check returns non-json format consistently.

Both issues fails the goal check and the conversation simulation will continue until the max_turn is hit.

This PR adds a retry logic when the LiteLLM completion call returns a temperature error and also adds response format to the goal check.

How is this PR tested?

• Existing unit/integration tests
• New unit/integration tests
• Manual tests

Manual Testing

Tested with databricks notebook: https://e2-dogfood.staging.cloud.databricks.com/editor/notebooks/2791734932961460?o=6051921418418893

Verified goal check works for databricks default model

Verified goal check works for databricks gpt-5 model

Verified goal check works for databricks claude sonnet model

Verfieid goal check works for databricks gemini 3 pro model

Does this PR require documentation update?

• No. You can skip the rest of this section.
• Yes. I've updated:
  • Examples
  • API references
  • Instructions

Release Notes

Is this a user-facing change?

• No. You can skip the rest of this section.
• Yes. Give a description of this change to be included in the release notes for MLflow users.

What component(s), interfaces, languages, and integrations does this PR affect?

Components

• area/tracking: Tracking Service, tracking client APIs, autologging
• area/models: MLmodel format, model serialization/deserialization, flavors
• area/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registry
• area/scoring: MLflow Model server, model deployment tools, Spark UDFs
• area/evaluation: MLflow model evaluation features, evaluation metrics, and evaluation workflows
• area/gateway: MLflow AI Gateway client APIs, server, and third-party integrations
• area/prompts: MLflow prompt engineering features, prompt templates, and prompt management
• area/tracing: MLflow Tracing features, tracing APIs, and LLM tracing functionality
• area/projects: MLproject format, project running backends
• area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev server
• area/build: Build and test infrastructure for MLflow
• area/docs: MLflow documentation pages

How should the PR be classified in the release notes? Choose one:

• rn/none - No description will be included. The PR will be mentioned only by the PR number in the "Small Bugfixes and Documentation Updates" section
• rn/breaking-change - The PR will be mentioned in the "Breaking Changes" section
• rn/feature - A new user-facing feature worth mentioning in the release notes
• rn/bug-fix - A user-facing bug fix worth mentioning in the release notes
• rn/documentation - A user-facing documentation change worth mentioning in the release notes

Should this PR be included in the next patch release?

Yes should be selected for bug fixes, documentation updates, and other small changes. No should be selected for new features and larger changes. If you're unsure about the release classification of this PR, leave this unchecked to let the maintainers decide.

• Yes (this PR will be cherry-picked and included in the next patch release)
• No (this PR will be included in the next minor release)