github llama.cpp · 3h ago · inference

llama.cpp b9990 adds Minimax2 eagle3 support and fixes nullptr

from English

The llama.cpp project released build b9990, which introduces support for the Minimax2 eagle3 model and resolves a nullptr issue within that implementation.

Added Minimax2 eagle3 support to the codebase.
Fixed a nullptr crash in the minimax2 EAGLE3 module.
macOS Apple Silicon builds with KleidiAI are disabled.
openEuler x86 builds are disabled.
Binaries are provided for macOS, Linux, Android, Windows, and UI across CPU, GPU, and specialized hardware backends.

Importance 1/3 Trust 1/3 llama.cpp Hugging Face Inference efficiency

Read original