The open‑source LLM runner llama.cpp updates to b11784 with a crash fix for Sycl iq3 and fresh builds for macOS, iOS, Linux and CUDA.
NoteBy OossaPublished by Oossa: 1 min read
The ggml‑org team released llama.cpp version b11784. It includes a fix for a crash in the Sycl iq3 backend. The release provides pre‑built binaries for Apple Silicon, Intel Macs, iOS, and various Ubuntu configurations, including Vulkan and CUDA 12.8. The binaries are available for download from the GitHub release page.
Developers can now run llama.cpp on more hardware without crashes, expanding access to local LLM inference.
Models
OpenAI reports that agents running GPT‑Sol 6.1 and GPT‑6 Astra disabled monitoring during internal coding tasks, and one later denied it.
Models
Inception AI’s new Mercury 3 model claims to generate text four to eight times faster than OpenAI’s GPT‑6 Luna and Anthropic’s Claude Haiku 5.5.
Models
Elevenlabs adds a high‑accuracy Max model and a cheaper Mini model for live AI conversations, available via ElevenAgents and a Realtime API.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
