Skip to content

Use Pi’s Responses provider for direct OpenAI runs - #420

Merged
asselstine merged 1 commit into
mainfrom
codex/pi-openai-responses
Nov 21, 2026
Merged

asselstine merged 1 commit into
mainfrom
codex/pi-openai-responses

Conversation

@asselstine

Copy link
Copy Markdown
Collaborator

Summary

  • Route the standard OpenAI API base URL through Pi’s built-in OpenAI provider and its Responses adapter. Third-party OpenAI-compatible URLs still use Pi’s Chat Completions adapter.
  • Resolve the saved integration credential through Factorize and preserve the configured context and output limits.
  • Exercise GPT‑6 Luna with High reasoning and a tool round trip in the bundled Worker using a mock OpenAI stream.

Validation

  • npm run test:pi-runtime passed on the debug VM. API TypeScript checking also passed.
  • Full VM validation reported a Phoenix telemetry test failure. Checking the result in isolation and completing the remaining validation.

@asselstine
asselstine marked this pull request as ready for review November 21, 2026 00:26
@asselstine
asselstine merged commit 5ede426 into main Nov 21, 2026
1 of 2 checks passed
@asselstine
asselstine deleted the codex/pi-openai-responses branch November 21, 2026 00:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant