The open‑source LLM runtime llama.cpp v0.1.0 adds GLM‑5.3‑Flash model support and several speed and memory fixes.
NoteOossaPublished by Oossa: 1 min read
The llama.cpp project released a new version on September 30, 2026. It adds support for GLM‑5.3‑Flash (also called GLM5‑Next), a newer language model architecture. The update also reduces compute buffer size, speeds up long‑context decoding, and fixes several bugs around token handling and GPU memory. The changes are aimed at developers who run LLMs locally on CPUs or GPUs.
The new features let users run the latest GLM models faster and with less memory on their own hardware.
Last updated: ·Markdown
Models
About 100 sites get paid for content used in AI Overviews, AI Mode and Gemini, but most earn under $1,000 and don’t know how payouts are calculated.
Models
OpenAI’s new GPT‑6.1 Sol keeps the $2‑$10 per‑million‑token rate while claiming stronger performance, undercutting Anthropic’s Claude and DeepSeek models.
Models
The new v4 and v4 Turbo voices cost $0.08 and $0.04 per 1,000 characters, with a promotional cut to $0.022 and $0.011 per 1,000 until Oct 12.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
