Skip to content

v6.2.0

Choose a tag to compare

View all tags
@Nayjest Nayjest released this 02 Jun 19:29
· 8 commits to main since this release
v6.2.0
ad5c26a
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

New Features

  • #138: Native support for using CLI commands to run LLM inference by @Nayjest in #139
  # === Example: Claude Code as inference backend ===
  LLM_API_TYPE=cli
  LLM_CLI="claude -p <request>"
# === Example: Gemini CLI as inference backend ===
LLM_API_TYPE=cli
LLM_CLI="gemini --skip-trust -p <request>"
  • Added compatibility with the OpenAI Python SDK v2+

Fixes & Minor Changes

  • Added an ApiPlatform.GOOGLE_AI_STUDIO check in the Google GenAI client to conditionally bypass Vertex AI–specific configuration (project, location, credentials, service account). This ensures the client is cleanly initialized with only an API key and vertexai=False when targeting Google AI Studio directly.
  • Bug fixes and refactoring in utils.resolve_callable, with improved exception handling
  • Minor refactoring of config validation

Internal SDLC Improvements

  • [Security / Maintainability] Enhanced test workflow by @akimochkinv in #136
    • Commit coverage badges to an orphan branch to avoid merge conflicts
    • Read-only permissions for test execution; secure commits made from a separate step with no custom code execution
  • AI code reviews with Gito:
    • Use Claude Opus 4.8
    • Include issues with confidence level ≥ 2
  • Documentation fixes

Full Changelog: v6.1.1...v6.2.0

Contributors

Nayjest and akimochkinv
Assets 2
Loading