Change8

v0.22.0-rc1

📦 ollamaView on GitHub →
1 features🐛 1 fixes🔧 2 symbols

Summary

This release introduces support for NVIDIA TensorRT Model Optimizer import within mlx and fixes an issue related to multi-regex BPE offset handling in the tokenizer. It also includes performance improvements by batching the sampler across multiple sequences in mlxrunner.

✨ New Features

  • mlx: Support NVIDIA TensorRT Model Optimizer import.

🐛 Bug Fixes

  • tokenizer: fix multi-regex BPE offset handling.

Affected Symbols