{"slug": "ornith-1-5-closes-the-self-improvement-loop-on-open-source", "title": "Ornith-1.5 Closes the Self-Improvement Loop on Open Source", "summary": "DeepReinforce released Ornith-1.5, an open-source AI model that closes the self-improvement loop by proposing its own training tasks, writing its own scaffolds, and generating its own RL rollouts without human curation. The flagship 397B variant scores 86.1 on Terminal-Bench 2.1, surpassing Claude Opus 4.8's 85.0, and is MIT licensed. The model's DeepSWE benchmark score jumped from 8 to 56.", "body_md": "DeepReinforce shipped Ornith-1.5 this week, and the Hacker News crowd landed on it fast. The new release extends its predecessor’s self-scaffolding approach into something more significant: a closed self-improvement loop where the model proposes its own training tasks, writes its own scaffolds, and generates its own RL rollouts — no human curation in the chain. The flagship 397B variant now scores 86.1 on Terminal-Bench 2.1, ahead of Claude Opus 4.8 at 85.0. It’s MIT licensed. You can run the 9B on Ollama right now. The Number That Matters: DeepSWE Went from 8 to 56 Benchmark tables in AI releases have […]\n\nThe post", "url": "https://wpnews.pro/news/ornith-1-5-closes-the-self-improvement-loop-on-open-source", "canonical_source": "https://byteiota.com/ornith-15-self-improving-open-source-agentic-model/", "published_at": "2026-08-19 18:09:59+00:00", "updated_at": "2026-08-19 18:57:25.494954+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "ai-products"], "entities": ["DeepReinforce", "Ornith-1.5", "Claude Opus 4.8", "Terminal-Bench 2.1", "DeepSWE", "Ollama"], "alternates": {"html": "https://wpnews.pro/news/ornith-1-5-closes-the-self-improvement-loop-on-open-source", "markdown": "https://wpnews.pro/news/ornith-1-5-closes-the-self-improvement-loop-on-open-source.md", "text": "https://wpnews.pro/news/ornith-1-5-closes-the-self-improvement-loop-on-open-source.txt", "jsonld": "https://wpnews.pro/news/ornith-1-5-closes-the-self-improvement-loop-on-open-source.jsonld"}}