feat(azure-open-ai): update model YAMLs [bot]#911
Conversation
|
/test-models |
Gateway test results
|
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit bb7a1a9. Configure here.
| - json_output | ||
| limits: | ||
| context_window: 400000 | ||
| max_input_tokens: 272000 |
There was a problem hiding this comment.
Removed context_window instead of updating it
Medium Severity
The context_window field was removed from gpt-5.4-mini-2026-03-17.yaml rather than being updated. All sibling models in the 5.4 family (gpt-5.4-2026-03-05.yaml, gpt-5.4-pro-2026-03-05.yaml) define context_window alongside max_input_tokens, and the new gpt-5-mini-2025-08-07-lite.yaml also includes it. When max_input_tokens was changed to 1050000, context_window likely needed to be updated (e.g., to 1050000) rather than deleted. Any consumer relying on context_window will now get no value for this model.
Reviewed by Cursor Bugbot for commit bb7a1a9. Configure here.


Auto-generated by poc-agent for provider
azure-open-ai.Note
Medium Risk
Changes provider model YAML configs that affect advertised token limits and per-region pricing; misconfiguration could impact routing, cost estimation, or request validation.
Overview
Adds a new Azure OpenAI model definition for
gpt-5-mini-2025-08-07-lite, including per-region costs, supported features/modalities, token limits, and exposed params (e.g.,max_tokens,reasoning_effort,verbosity), markedactivewiththinking: true.Updates
gpt-5.4-mini-2026-03-17limits by raisingmax_input_tokensfrom 272k to 1,050,000 and dropping the explicitcontext_windowentry.Reviewed by Cursor Bugbot for commit bb7a1a9. Bugbot is set up for automated code reviews on this repo. Configure here.