ABUS Gives Omarchy Agents Their Own Mailbox and Message Bus
Here’s a fresh Omarchy plugin for anyone running more than one AI agent. Following DHH’s idea of networking agents across Omarchy machines, corelumen released ABUS, an…
Category
AI & Agents
Follow AI & Agents via RSSWhat’s New in Hermes Agent 0.22.3 and 0.22.4
Hermes Agent shipped two stable releases on October 21, the first since 0.22.2. 0.22.3 A big roll-up of about 620 merged PRs since 0.22.2. Full curated…
What’s New in Claude Code 2.1.307 and 2.1.308
Two more Claude Code releases landed October 21 and 22. 2.1.307 highlights /claude-api api-version moves your Claude API code to API version 2026-10-21. A new claude_code.safety_event…
Hugging Face’s Carbon-X Lets You Chat With DNA, Locally
Here’s a different kind of open model. Hugging Face released Carbon-X on Thursday, a 14 billion parameter model you can talk to about DNA, and it’s…
Ollama 0.40.8 Brings Flint, a 40B Coding Model for Agents
Ollama 0.40.8 is out, and the headline is Flint, Recursive’s coding model, now on Ollama. About Flint An Apache 2.0, 40B mixture-of-experts model with about 2.5B…
LiteLLM 1.107.0 Was Trojaned: Check for It and Rotate Your AI Keys
Attackers compromised LiteLLM’s release process on Wednesday and pushed a malicious 1.107.0 to PyPI and Docker. It steals cloud and AI credentials and drops a remote…
Anthropic Launches Claude Duet, a Proactive Agent That Reaches Out First
Anthropic just introduced Claude Duet, its answer to Meta’s Muse and OpenAI’s Dots. It’s a proactive agent that “reaches out before you ask.” What it does…
NVIDIA Woos Europe With More Open-Weight Models
At GTC Berlin, NVIDIA added more open-weight models to its library, pitched at Europe’s appetite for sovereign AI. What’s new Refreshed Nemotron Retriever embedding, parsing and…
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…
IBM’s CEO: Open-Weight Models Are Now ‘Irresistibly’ Good and Cheap On-Prem
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…