On-Device & Local Inferenceggml project

ggml project introduces the GGUF (GGML Universal File) format

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

GGUF (GGML Universal File) format introduced — stores tensors and metadata in a single file, is mmap-compatible, and extends with backward compatibility through key-value metadata

Why it matters

The de facto standard format for deploying local LLMs

Source — Primary sources first; secondary reports are labelled “Press”

Pressen.wikipedia.orghttps://en.wikipedia.org/wiki/Llama.cpp

Added in Week 35, 2026 · See that week

On the map — ⑲ On-Device line

  1. Previous stopllama.cpp
  2. This recordggml project introduces the GGUF (GGML Universal File) format
  3. Next stopPhi-3
See it on the map

Nearby on the same axis

Georgi Gerganov starts developing llama.cppPressMicrosoft Research publishes Phi-3 technical reportPaperFrantar et al. present GPTQPaperApple unveils the Apple Intelligence on-device foundation modelOfficial