The llama.cpp project has released version b12407, which includes a workaround for Metal GPU acceleration related to the LFM3 model. This update, assisted by Claude Opus 5.5, focuses on optimizing tensor operations for norm fusion and updating Metal fusion baselines for LFM3. AI
IMPACT Improves performance for users running LFM3 models on macOS with Metal GPUs via llama.cpp.
RANK_REASON This is a software release for a specific tool (llama.cpp) that includes optimizations for a particular model (LFM3) and hardware acceleration (Metal), rather than 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 →