Draft: Fix loading order of API key customization variable#3
Draft
stevemolitor wants to merge 28 commits intomainfrom
Draft
Draft: Fix loading order of API key customization variable#3stevemolitor wants to merge 28 commits intomainfrom
stevemolitor wants to merge 28 commits intomainfrom
Conversation
Mistral will return an error if no model is provided in the request.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
WIP fix for #2.
The
robby-openai-api-keycustomization variable was not getting loaded early enough, causing any value set in the user's init file to get overridden when a robby command was invoked and autoloaded. Therobby-openai-api-keycustomization variable was declared in a file that was not loaded early. To fix, move the custom variable declaration to `robby-customization.el', which is eagerly loaded.Also, fetching available models is not working for some reason with mistral or togetherai. As a temporary fix (in this branch only), comment out the "m - model" option in the robby options transient, to avoid triggering that request. I'll figure out a better solution before merging this.
Finally, I noticed that conversation history is not working with mistral. There's an extra "assistant" message in the payload that messes it up (but does not bother OpenAI), should be an easy fix.
I'm putting this PR out there now for early feedback.