●Change8

v0.17.1-rc2

📦 ollamaView on GitHub →
✨ 3 features🐛 2 fixes

Summary

This release introduces support for the nemotron architecture and includes several performance and logging improvements, particularly for MLX-based operations. It also updates underlying MLX-C bindings.

Migration Steps

  1. Update mlx-c bindings to version 0.5.0.

✨ New Features

  • Added support for the nemotron model architecture.
  • Implemented capability-based detection for web search in the UI.
  • Added upgrade configuration option to the settings page.

🐛 Bug Fixes

  • Reduced log noise and fixed duplicate log prefixes in mlxrunner.
  • Stopped defaulting to affine quantization for models that are not quantized.