ggml

The latest AI and IT news about ggml, newest first, with key points summarized by DevPick's AI.

Products, companies and technologies often covered in the same articles as ggml.

Latest articles and summaries (3)

PC Watchby 宇都宮 充11/16/2026
AI summary

The v1.0.0 of the C/C++ open-source LLM inference engine llama.cpp was released on November 13, marking its first major update in about three and a half years.

  • •Version 1.0.0 of llama.cpp, built with C/C++, was released on November 13
  • •It is the first major version release after approximately three and a half years of development
  • •The related tensor library ggml was also updated to version 1.0.0
#LLM / Model Trends#Generative AI Tools & Use
gihyo.jp11/14/2026
AI summary

llama.cpp released its first major version 1.0.0, marking a shift to explicitly indicate API and ABI incompatible changes.

  • •The development team released the first major version 1.0.0 on November 13, 2026.
  • •It functions as a C/C++ library for executing inference of large language models.
  • •The tensor library ggml, which supports the calculations, also reached version 1.0.0 simultaneously.
#LLM / Model Trends#Generative AI Tools & Use
Hacker NewsHN submitter: cobychapple11/14/2026
AI summary

ggml 1.0.0 removes deprecated public APIs and legacy DIAG_MASK operations, adds configurable YaRN pruning and FP8 support, and includes extensive performance work across multiple back‑ends, including low‑row quantization.

  • •Deprecated public APIs and legacy DIAG_MASK operations are removed.
  • •Configurable YaRN pruning and FP8 support are added.
  • •Extensive performance work across multiple back‑ends, including low‑row quantization, is included.
#LLM / Model Trends#Generative AI Tools & Use