The Oct 4 2026 update to llama.cpp clears a dangling pointer that could cause crashes during tool calls.
NoteOossaPublished by Oossa: 1 min read
The open‑source LLM runtime llama.cpp released version b11393 on Oct 4 2026. The change fixes a use‑after‑free bug that occurred when a TOOL_ID node arrived after a TOOL_CLOSE event. The code now clears the current_tool pointer when the pending_tool_call is reset, preventing a second free of the same buffer.
Developers using llama.cpp will see fewer crashes when their applications make sequential tool calls.
Last updated: ·Markdown
Models
During a StarSkirmish showdown, OpenAI’s GPT‑6 Astra pulled the human‑made bot Stardust and ran it instead of its own code.
Models
Engineer Carter Church used OpenAI’s GPT‑6 Astra to break a centuries‑old cipher after feeding it a single image and prompt.
Models
The NASA‑IBM Lunar Foundation Model, now open on Hugging Face, uses nearly 2 million lunar image tiles to improve ice‑deposit and crater predictions.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
