Starstruk Logo MarkStarstruk

v9.8.0

a16z-infra/JungleGymv9.8.0Apr 29, 2026by davidmezzetti

AI Summary

This release is primarily a bug fix, security, and compatibility update addressing issues with LiteLLM, llama-cpp-python, and Model2Vec, while adding support for last token pooling and streaming Labels pipeline.

Key Highlights

  • Fixed a security issue with LiteLLM.
  • Fixed compatibility issues with latest llama-cpp-python and Model2Vec.
  • Added support for last token pooling.
  • Added support for streaming Labels pipeline.

New Features

  • Last token pooling support
  • Streaming Labels pipeline support
  • Safeopen parameter for Textractor
  • Authorization tokens support for MCP endpoints

Full Release Notes

**This release is primarily a bug fix, security and compatibility release.**

See below for full details on the new features, improvements and bug fixes.

New Features
--------------------------
- Add support for last token pooling (#1072)
- Add support for streaming Labels pipeline (#1075)
- Add safeopen parameter for Textractor (#1077)

Improvements
--------------------------
- Support Authorization tokens with MCP endpoints (#899)
- Update embeddings documentation to mention loading ONNX vector models (#1071)
- Performance Improvement: Use JOIN when JSON fields being used (#1078)
- Limit template scope with Agents (#1080)

Bug Fixes
--------------------------
- Security issue with LiteLLM (#1065)
- Issue with latest version of llama-cpp-python (#1066)
- Issue with latest version of Model2Vec (#1067)
- UnsupportedProtocol: Request URL is missing an 'http://' or 'https://' protocol (#1073)