Repository navigation
Normalize model metadata for OpenAI-compatible providers - #621
Merged
Merged
Conversation
Prevents embedded first-party capabilities from leaking into proxy and custom-provider requests. Preserves remote overrides for the supported first-party path while using conservative fallback metadata elsewhere.
Avoids treating arbitrary provider model endpoints as authoritative Codex metadata sources. Keeps refreshes available for Codex-backed and command authentication flows that provide the expected schema.
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
/modelsdiscovery to Codex-backend or command-backed provider authentication.Motivation
OpenAI-compatible proxies can expose the same model slugs as first-party OpenAI, but they do not necessarily support OpenAI-only transport and request semantics. This change keeps those providers on ordinary Responses behavior instead of silently adopting provider-specific metadata from embedded OpenAI model definitions.
Remote model discovery is also gated consistently so bare OpenAI-compatible
/modelsresponses are not treated as authoritative CodexModelInfocatalogs.The 5.6 models advertise the same 1M context capability as GPT-5.4 in this code-rs architecture, so they now participate in the existing context-mode selection rather than being capped at the 272K default.
No gateway-specific addresses, model names in code paths, or user-facing configuration knobs are added.