NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native models developed to explore this path through agentic post-training. Our system combines a heterogeneous model pool with intelligent routing, recording the
Reading in Plain English mode — toggle below to switch
Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native models developed to explore this path through agentic post-training. Our system combines a heterogeneous model pool with intelligent routing, recording the
This work introduces a novel approach to recursive self-improvement for AI agents, leveraging agentic post-training and intelligent routing. Understanding this paradigm is crucial for ML engineers aiming to build more autonomous and continuously improving AI systems.