Repository navigation
feat(openai): update model YAMLs [bot] - #5547
models-bot[bot] wants to merge 1 commit into
Conversation
SDK live feature checksResult: failed (125/485 scenarios) Informational — does not block merge.
Failures
|
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort 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 98eb3a1. Configure here.
| input_cost_per_token_batches: 1e-7 | ||
| output_cost_per_token: 0.0000012 | ||
| output_cost_per_token_batches: 6e-7 | ||
| priority_pricing: |
There was a problem hiding this comment.
Dropped still-valid priority pricing
Medium Severity
This update deletes priority_pricing from gpt-5.6-luna and gpt-5.1-2025-11-13, but OpenAI still publishes Fast/Priority rates that match the removed values. Remaining tiered_pricing on luna is standard long-context, not priority. Fast/service_tier cost estimates will fall back to standard prices and undercount spend.
Additional Locations (1)
Reviewed by Cursor Bugbot for commit 98eb3a1. Configure here.


Auto-generated by poc-agent for provider
openai.Note
Low Risk
Catalog-only YAML updates; main behavioral risk is consumers treating newly deprecated models differently or losing priority-tier pricing on the two models where that block was removed.
Overview
Updates OpenAI provider model metadata to reflect upcoming sunsetting: gpt-4.1, gpt-4o, gpt-4o-mini, gpt-5-mini, gpt-5.1, gpt-4o-mini-tts, and tts-1 variants move from
activetodeprecated, with newisDeprecated,deprecationDate, andretirementDatefields (dates vary by model family).Token and character costs are unchanged in value but rewritten from decimal literals to scientific notation (e.g.
2e-6instead of0.000002). A few files also droppriority_pricingblocks (gpt-5.1-2025-11-13,gpt-5.6-luna) and normalizeremoveParamsentries (unquotedn).Reviewed by Cursor Bugbot for commit 98eb3a1. Bugbot is set up for automated code reviews on this repo. Configure here.