Tag
Offloop launches persistent AI agents that carry context across cycles to reduce recurring ops overhead for small teams, claiming state-of-the-art performance on GDPval ahead of Claude code and Codex.
The article argues that persistent AI agents require snapshot-based recovery of their entire environment, not just memory, and introduces Mantlecore AI's VM snapshot feature for reversible updates.
A tweet speculates about an OpenAI Agent phone concept that runs persistent AI agents 24/7 without traditional apps.
Nous Research and NVIDIA have independently converged on the same architecture for persistent AI agents that live on servers and improve daily, marking a shift from coding copilots to autonomous server-side agents.
This article introduces iLands, a project exploring autonomous AI agents as persistent social and economic entities with memory and real-world context, rather than mere task executors.
AutoLab is a new benchmark evaluating 17 frontier models on 36 expert-curated long-horizon tasks (system optimization, model development, CUDA kernels, puzzles), finding that persistence—not initial attempt quality—is the dominant predictor of success. Claude-opus-4.6 led all categories, while most other models terminated prematurely or exhausted budgets with minimal progress.
duet-agent is a harness for persistent AI agent jobs that outlive a single chat session, enabling long-running, resumable tasks with cross-session memory and observable state through relay turns.