Starstruk Logo MarkStarstruk

v9.1.0

neuml/txtaiv9.1.0Nov 4, 2025by davidmezzetti

AI Summary

This release focuses on vector 'un-databases', allowing vectors to be stored using NumPy, Torch, and GGUF formats from llama.cpp. It also introduces quantization and Safetensors storage support.

Key Highlights

  • Add GGML ANN support
  • Add quantization support to Torch ANN
  • Add support for Safetensors storage to NumPy/Torch
  • Add GraphRAG example notebook

New Features

  • GGML ANN support
  • Torch ANN quantization
  • Safetensors storage support for ANN backends

Full Release Notes

**This release focuses on vector "un-databases" - store vectors with NumPy, Torch and even GGUF from llama.cpp!**

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

New Features
--------------------------
- Add GGML ANN (#975)
- Add quantization support to Torch ANN (#971)
- Add support for Safetensors storage to NumPy/Torch ANN backend (#969)
- Add example notebook demonstrating how to access low level vector APIs (#973)
- Add GraphRAG example notebook with wikipedia and gpt-oss (#967)

Improvements
--------------------------
- Update LLM stripthink parameter to support gpt-oss patterns (#966)
- Pin mcpadapt==0.1.16 until upstream issue resolved (#970)

Bug Fixes
--------------------------
- Not possible to load a False ANN setting (#976)