feat(azure-open-ai): update model YAMLs [bot] - #2079
Conversation
|
/test-models |
Gateway test results
|
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort and found 3 potential issues.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit ff07bc7. Configure here.
| input: | ||
| - text | ||
| output: | ||
| - text |
There was a problem hiding this comment.
Incomplete chat-latest modalities
Medium Severity
The new modalities.input list only includes text, but gpt-chat-latest is documented as multimodal with image input, and the prior Azure snapshot also lists image and pdf. Recording text-only support can incorrectly block multimodal usage that depends on this metadata.
Reviewed by Cursor Bugbot for commit ff07bc7. Configure here.
| input: | ||
| - text | ||
| - image | ||
There was a problem hiding this comment.
Stale terra retirement date
Medium Severity
retirementDate for gpt-5.6-terra-2026-07-09 remains 2027-07-09, while the same-version luna and sol models in this PR were updated to 2028-01-11. Before this change all three shared the earlier date, so terra is now inconsistently marked as retiring much sooner.
Additional Locations (2)
Reviewed by Cursor Bugbot for commit ff07bc7. Configure here.
| input: | ||
| - text | ||
| - image | ||
There was a problem hiding this comment.
Inconsistent sol input limit
Medium Severity
max_input_tokens for gpt-5.6-sol stayed at 922000, while this PR raised luna and terra from 922000 to 1050000 to match their shared context_window. Sol now uniquely under-reports the allowable input size for the same 1.05M context family.
Additional Locations (2)
Reviewed by Cursor Bugbot for commit ff07bc7. Configure here.
1 participant
Auto-generated by poc-agent for provider
azure-open-ai.Note
Medium Risk
Incorrect token costs or limits in these YAMLs would skew billing/quoting and client-side validation for production model routes.
Overview
Auto-generated refresh of Azure OpenAI model metadata for four YAML definitions.
gpt-5.6-luna and gpt-5.6-terra get lower per-token and tiered prices (roughly ~80% down on luna, ~20% on terra) across
datazone_us,datazone_eu, and*. Luna and terra also raisemax_input_tokensfrom 922,000 to 1,050,000 (full context window).gpt-5.6-luna, gpt-5.6-sol, and gpt-5.6-terra add PDF under input modalities. gpt-5.6-luna and gpt-5.6-sol extend
retirementDatefrom 2027-07-09 to 2028-01-11.gpt-chat-latest-2026-06-24 gains an explicit modalities section (text in / text out); limits and other fields are unchanged in this diff.
Reviewed by Cursor Bugbot for commit ff07bc7. Bugbot is set up for automated code reviews on this repo. Configure here.