fix: set api:openai-completions per-model for non-anthropic/ models - #6
Open
GiorgioBarresi wants to merge 2 commits into
Open
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
The Requesty docs recommend configuring the provider with
api: "anthropic-messages". This works correctly for directanthropic/claude-*models, but breaks every other model (Azure, Vertex Gemini, Mistral, etc.) because Pi sends those requests in Anthropic Messages format, which those providers reject with a 422.Fix
In
discoverModels, add a per-modelapi: "openai-completions"override for any model whose ID does not start withanthropic/. Direct Anthropic models continue to inherit the provider-levelanthropic-messagessetting (preserving native prompt caching), while all other models get the correct OpenAI-compatible format.updateModelsJsonis also updated to persist theapifield so the override survives a/requesty-models-sync+/reloadcycle.