The open‑source model server adds cloud‑usage APIs, fixes Windows bugs and streamlines CLI onboarding.
NoteBy OossaPublished by Oossa: Last updated: 1 min read
Ollama posted version 0.40.1‑rc0 on 8 October 2026. The update adds a proxy for cloud‑usage and balance APIs, removes the account step from the command‑line onboarding, and fixes a Windows bug that let the LLaMA model read past 2 GiB. It also avoids Windows symlinks and drops a metal‑memory patch now upstream. New contributors made three first‑time commits.
Users of Ollama on Windows can run larger models without crashes, and new users can start the CLI faster.
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.
