NeoHorse-1: Recursive self-improvement via agentic post-training with routing harness

TokenRhythm · hf · 2026-09-09

NeoHorse-1, released on Hugging Face, pursues recursive self-improvement through agentic post-training: it combines intelligent routing, structured feedback loops, and curriculum-based distillation to improve model capabilities across agent benchmarks.

Related event: NeoHorse-1 Paper Proposes Recursive Self-Improvement via Agentic Post-Training(2 posts)→

Original post →

More from Research

Research channel →