The llama.cpp project has released version b12233, introducing the capability to load external PLE (Portable Low-level Executable) files. This update allows for multi-shard loading, where external files can be appended as extra shards, bypassing standard split checks. The release also includes changes to command-line arguments, making `--ple` an alias for `--shard`, and refactors model loading to utilize a unified path for multiple shard locations. AI
IMPACT Enhances the flexibility of model loading for the llama.cpp tool.
RANK_REASON This is a software release for a specific tool, not a frontier model release or significant industry event.
Read on llama.cpp — Releases →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →