Update: Meta’s 816B Muse Spark Open Runs on Three Old Omarchy Laptops and a Mac Studio
Update to my earlier post, Meta’s Muse Spark Open: 816B Open Weights, Two Mac Studios. When Meta released the weights, Mark Zuckerberg said the model “runs…
etherninja.org Saving the Internet One Packet at a time!
Subscribe via RSS
Tag
Update to my earlier post, Meta’s Muse Spark Open: 816B Open Weights, Two Mac Studios. When Meta released the weights, Mark Zuckerberg said the model “runs…
Moose is a GNOME-native AI chat app built on GTK4 and libadwaita. It started as a local chat app and has grown into cloud providers too.…
The Swiss AI team published full release notes and a technical report for Apertus 1.5, the update to its fully open 8B and 70B models. The…
Coding agents burn a lot of tokens just figuring out where to look. LightOn’s answer is ColGrep-Agent-2B, an open 2-billion-parameter code-search agent that runs locally between…
Update to my earlier post, Ollama 0.40.5: Nemotron 3.5 Super and Bolt for Local Agents. Ollama has moved on twice since then. Ollama 0.40.6 (stable, Oct…
Ollama 0.40.5 shipped on October 14, 2026, and the headline is NVIDIA’s Nemotron 3.5 family landing for local (and cloud-adjacent) runs. Super is the big sibling;…
October 16, 2026: NVIDIA says RTX Spark laptops are available today from ASUS, Dell, HP, Lenovo, Microsoft, and MSI. The pitch is Blackwell RTX + CUDA/AI…
Meta released the weights for Muse Spark Open on October 15 and is calling it the most capable US open-weight model to date. The specs: natively…
Kernel backports are where quiet bugs hide. NVIDIA’s Boro, presented around LPC in Prague, is an open-source Rust CLI that uses local AI to help validate…
Quiet tooling releases matter more than model launch theater. On October 8, Ollama 0.40.2 landed as stable — MLX as the default on Apple Silicon, out…