The open‑source llama.cpp library v b11594, released Oct 13 2026, lets multiple GPUs share host memory and adds official Windows builds.
NoteBy OossaPublished by Oossa: 1 min read
The ggml‑org team released llama.cpp version b11594 on Oct 13 2026. The update adds Vulkan support for multiple devices to share GPU‑visible host memory. It also includes a clean Windows binary, fixes for null pointers, and minor cleanup. Pre‑built packages are provided for macOS, iOS, Linux (CPU and Vulkan) and now Windows.
Developers can now run llama.cpp models on several GPUs at once and on Windows PCs, expanding hardware options.
Models
Max routes requests to the most suitable model, cutting cost by up to 55% while improving performance in chat, code and work tasks.
Models
Google’s Gemini 4 Argon model is now in Devin Desktop and CLI at 75% off, and scores top on the FrontierCode 1.1 benchmark.
Models
The new Gemini 4 Argon AI model is now available globally via API and Google apps, priced at $2 per million input tokens.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
