fix(core): align ChatGPT OAuth context limits - #39541
Closed
magoz wants to merge 1 commit into
Closed
Conversation
github-actions
Bot
added
the
needs:compliance
Contributor
|
The following comment was made by an LLM, it may be inaccurate: Potential Duplicate FoundPR #38987: fix(core): honor Codex input limits Why it might be related:
|
github-actions
Bot
removed
the
needs:compliance
Contributor
|
Thanks for updating your PR! It now meets our contributing guidelines. 👍 |
Author
|
Closing in favor of the existing maintainer-requested V2 fix #38987. That PR targets the actual |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Issue for this PR
Closes #38851
Type of change
What does this PR do?
Applies V1's ChatGPT OAuth-specific GPT-5.5/5.6 context limits in V2. Public API-key limits and other models remain unchanged. The catalog reloads when the OpenAI connection changes.
ChatGPT's Codex backend exposes smaller windows than Models.dev's public API catalog. Without this overlay, provider overflow recovery compacts GPT-5.6 Sol near 372k while the TUI reports about 35% of 1.05M.
Catalog inspection reads stored credential type without refreshing tokens, so token expiry or refresh failure cannot alter model metadata.
How did you verify your code works?
Screenshots / recordings
Not a UI change.
Checklist