The open‑source llama.cpp library released version b11462 on Oct 7 2026, adding a SYCL code cleanup and binary packages for macOS, iOS, Linux and CUDA.
NoteBy OossaPublished by Oossa: 1 min read
ggml‑org released llama.cpp version b11462 on 2026‑10‑07. The update cleans up SYCL‑based attention buffers, a low‑level graphics‑compute interface. It also ships pre‑built binaries for macOS (Apple Silicon and Intel), iOS, and several Ubuntu flavors, including CPU, Vulkan and CUDA 12.8 builds. The binaries are available for download from the GitHub release page.
Developers can now run llama.cpp locally on more platforms without building from source, speeding up experimentation.
Models
The company’s new three‑agent Claude‑based reviewer found 73.43% of core‑claim errors, far above the previous best of 14.81%.
Models
Mistral's new trillion‑parameter model scores 38 on an independent index, but costs more per task than competing open models.
Models
OpenAI reported two influence campaigns that used AI‑generated content to plant false stories in real media and blocked the involved accounts.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
