aiminute. ← All AI news
Tools AI Minute Newsroom 2026-11-14

The engine under many local AI apps finally reached version 1.0.

The engine under many local AI apps finally reached version 1.0.

The ggml team released version 1.0 of its library on Friday, the math engine inside the popular llama.cpp. The release drops old interfaces and adds FP8, a compact number format that newer models use. Large models that wake only part of their weights now run faster on Nvidia and Apple chips.

Why it mattersIf you run models on your own computer, the apps you use are likely to get these speed-ups. Developers who rely on the removed old functions have to update their code first.
#Coding

✓ Verified · 3 sources

WhatsApp X Telegram
Read in the app — free, in 9 languages

Related stories

Canva adds a switch that turns off most of its generative AI.
2026-11-19
Ruby fixes a critical flaw that AI tools helped uncover.
2026-11-19
Meta staff fear its Muse agent mixed up users' bank details.
2026-11-19
Google opens its Genie 4 world generator to top-tier subscribers.
2026-11-19
Four security teams used AI to find the same critical Nginx flaws.
2026-11-19