Should Local AI Agents Live in a Sandbox?
Two stories this week got me thinking about this. Perplexity Computer now runs fully local on a Mac with tools in isolated sandboxes, and NVIDIA wants…
etherninja.org Saving the Internet One Packet at a time!
Subscribe via RSS
Tag
Two stories this week got me thinking about this. Perplexity Computer now runs fully local on a Mac with tools in isolated sandboxes, and NVIDIA wants…
On IBM’s Q3 earnings call, CEO Arvind Krishna made the case that open-weight models now make total sense for running AI on your own hardware. What…
Mistral just dropped Voxtral 3, a new family of speech models plus a voice agent SDK that listens, thinks, calls tools and talks back in one…
Perplexity says Perplexity Computer, its agent that does tasks for you, can now run fully locally on Apple silicon Macs. What’s new It runs on Qwen…
AMD shipped GAIA 0.25, its open-source project for building and running local AI agents on Ryzen AI hardware. It started as “Generative AI Is Awesome” and…
Vercel’s AI Gateway numbers show open-weight models are winning on usage and catching up on spending. Open-weight models processed 71.5% of gateway tokens, and the majority…
DeepSeek dropped DeepSeek-V4.1-Flash-1017 on Hugging Face, a “minor update” to V4.1 Flash with big gains in reasoning, coding, agentic tasks and vision. Weights are under the…
The M6 Mac mini is on its way to running Omarchy. Joshua Warren of the Omarchy M team posted on October 18: “Installed a bootloader, got…
Nous Research pushed Hermes Agent 0.22.2 on October 15 (still the latest release), and the notes are refreshingly honest: 0.22.1 was built from the wrong commit.…
A new US lab called Reflection just shipped its first model, and it’s open. The hook: Ollama is still running it free for a limited time,…