add support for OpenAI-compatible LLM providers#2113
Open
tequeter wants to merge 2 commits into
Open
Conversation
Add OpenAICompatibleProvider, to make use of local models and alternative providers. OpenAICompatibleProvider is distinct from the existing OpenAIProvider to preserve backwards compatibility (most notably the completion tokens calculation). If compatibility was not an issue, OpenAIProvider and XAIProvider could be replaced by OpenAICompatibleProvider + configuration. Not included: cost tracking for the new provider. Co-authored-by: Codex <noreply@openai.com>
Add a WEBFEED_MODEL setting. Keep the current "haiku" as default. Co-authored-by: Codex <noreply@openai.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR allows configuring alternative LLM providers (such as OpenRouter or local inference) for:
Other AI features would require larger changes, as they are not using the
LLMProviderclass and registry.More details in individual commit messages.
Samuel, do you agree with the intent? Are there things you would have preferred to see done differently, in this PR or a future one?