llama.cpp adds MiniCPM5 tokenizer support in update b9354

Core summary

One sentence to understand this update

The latest llama.cpp release b9354 introduces support for the MiniCPM5 tokenizer, including its pre-tokenizer hash and regex handling.

Impact & opportunity

What this could mean

Developers can now leverage MiniCPM5 models within llama.cpp, expanding options for local LLM inference.