Repository navigation
Support user-provided AI providers and ChatGPT plans - #63
Merged
Merged
Conversation
darthvader58
marked this pull request as ready for review
November 4, 2026 22:45
Collaborator
|
Two fixes needed before merging:
|
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #24
People can now use TATC AI without ASU credentials or a VPN, by adding their own API key or, on a local install, connecting a ChatGPT plan. ASU Voyager and CreateAI stay available.
What changed
user_providedstorage, on both local and deployed stacks.apps/chatgpt-bridgeservice using OpenAI's Sign in with ChatGPT for locally hosted open-source apps. It holds the OAuth tokens and translates LibreChat's Chat Completions requests into the Responses requests the plan route requires. OpenAI requires an interest form for hosted apps, so the deploy stack leaves it out.docs/providers.mdcovers setup, eligibility, limits, billing, and which steps need the VPN.Verification
Done with a real ChatGPT plan, in the real LibreChat UI:
127.0.0.1:3080;localhost:3080was checked at the status endpoint only.Done in the real LibreChat UI (v0.8.7), using test accounts:
Not done yet:
Existing issues found, not changed here
empty_messagesbefore any request is sent, because LibreChat assumes a small context window and the tool definitions use about 44k tokens. Other model specs that LibreChat does not recognize may needmaxContextTokenstoo. Might want to check on this ^