Missing regression coverage for @ai-sdk/openai Responses default on custom providers

[vegerot]

Note

This issue was written by OpenCode on behalf of the contributor.

Description

Custom providers that use @ai-sdk/openai currently rely on the SDK default selected by languageModel(). OpenCode has explicit Responses handling for the built-in openai provider, but without a regression test on the custom-provider path, contributors cannot tell from the test suite whether that fallback will stay on the Responses API.

This guarantee matters for me and other OpenCode developers because provider work regularly involves custom provider IDs, proxy endpoints, and transport-specific behavior. If languageModel() ever stopped resolving to the Responses path for @ai-sdk/openai, custom OpenAI-backed providers could silently drift onto the wrong protocol and break integrations in ways that are hard to catch during review.

Plugins

None

OpenCode version

Current dev branch

Steps to reproduce

1. Configure a custom provider that uses @ai-sdk/openai.
2. Rely on the default sdk.languageModel(...) path instead of the built-in openai provider override.
3. Observe that the previous test suite did not explicitly guarantee this path stays on Responses.

Screenshot and/or share link

N/A

Operating System

macOS

Terminal

Ghostty

[github-actions]

This issue doesn't fully meet our contributing guidelines.

What needs to be fixed:

• The issue explicitly states it was written by OpenCode (an AI). AI-generated walls of text are not acceptable per our contributing guidelines. Please rewrite this in your own words, describing the actual problem you encountered and why this regression coverage matters to you specifically.

Please edit this issue to address the above within 2 hours, or it will be automatically closed.

---

This issue might be a duplicate of existing issues. Please check:

• [BUG] options.reasoningEffort not passed to API calls for custom @ai-sdk/openai provider #20815: Custom @ai-sdk/openai provider behavior — options not forwarded (closed, but closely related to the same custom provider + Responses API path)
• [FEATURE]: enable responses API in openai compatible provider #9628: Feature request to enable Responses API in OpenAI-compatible providers (same underlying concern about which protocol custom providers use)

If you believe this was flagged incorrectly, please let a maintainer know.

[github-actions]

This issue has been automatically closed because it was not updated to meet our contributing guidelines within the 2-hour window.

Feel free to open a new issue that follows our issue templates.