Mellum2.1: JetBrains’ Fast Open Coding Model
JetBrains shipped Mellum2.1 on October 8 — a 12B MoE coding model with about 2.5B active parameters, Apache 2.0, trained with RL in real repositories. It…
etherninja.org Saving the Internet One Packet at a time!
Subscribe via RSS
Tag
JetBrains shipped Mellum2.1 on October 8 — a 12B MoE coding model with about 2.5B active parameters, Apache 2.0, trained with RL in real repositories. It…
Most “agent sandboxes” are polite suggestions. Garuda (announced October 8) takes the grown-up route: Seatbelt on macOS, bubblewrap on Linux, no network by default, an audit…
Two Mac-shaped rumors showed up in the same OmarchyCentral roundup. First, X user StudioXRadio claimed Omarchy runs fully on an M4 MacBook Air, with a repo…
Distros get sticky when the app ecosystem feels native. OmarchyCentral’s October community roundup highlighted three: OmaEar (a podcast player that follows your theme), Omastrator 0.2.3 (a…
On October 8, 2026, Omarchy announced SpaceXAI as a Founding Corporate Patron — about $1.5M in Grok API tokens aimed at Omabot PR review. Official numbers…
Keyboard defaults start religious wars. OmarchyCentral reported on October 11 that the next Omarchy release will leave Caps Lock as Caps Lock again, and move compose…
A day after the SpaceXAI news, Omarchy posted that Namespace joined as a Distinguished Corporate Patron — $100K per year in compute for three years, aimed…
Model shopping is a trap if you chase every launch post. GenAI Brief’s October 2026 open-weight roundup is a useful shortlist: Gemma 4, gpt-oss, Mistral Medium…
DeepSeek’s agent runtime grew up. Harness v0.2 (developer preview, late September 2026) ships the first official desktop installers for Apple-silicon macOS and 64-bit Windows. Same MIT-licensed…
Magnitude is not another chat window. It is an open-source inference engine for local agents that tunes kernels on your hardware before a model runs. The…