Skip to content

fix(provider): preserve abliteration prompt cache reuse - #159

Open
mikemikimike wants to merge 1 commit into
CyberStrikeus:mainfrom
mikemikimike:fix/abliteration-cache
Open

mikemikimike wants to merge 1 commit into
CyberStrikeus:mainfrom
mikemikimike:fix/abliteration-cache

Conversation

@mikemikimike

Copy link
Copy Markdown

Summary

  • Enable prompt cache routing hints for abliteration.ai OpenAI-compatible models.
  • Keep the first-turn prompt prefix stable by skipping synthetic tool-call examples for this provider.
  • Preserve the existing setCacheKey: false opt-out.

Fixes #142

Issue: #142

Testing

  • bun test test/provider/transform.test.ts --test-name-pattern 'abliteration|setCacheKey'
  • bun typecheck

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.

No input-caching for abliteration.ai

1 participant