{"slug": "new-agentic-coding-sota-models", "title": "New agentic coding SOTA models", "summary": "Ornith-1.0, a family of open-source LLMs specialized for agentic coding, achieves state-of-the-art performance on multiple coding benchmarks including SWE-Bench and Terminal-Bench. The models range from 9B to 397B parameters and are released under the MIT license for commercial and research use.", "body_md": "Aloha! 🌺 Meet Ornith-1.0, a family of open-source LLMs specialized for agentic coding.\nOrnith-1.0 spans the full parameter sizes including 9B Dense, 31B Dense, 35B MoE, and 397B MoE. It achieves state-of-the-art performance among open-source models of comparable size on coding benchmarks including:\n✅Terminal-Bench 2.1(77.5)\n✅SWE-Bench(82.4 on verified, 62.2 on pro, 78.9 on Multilingual)\n✅NL2Repo(48.2)\n✅SWE Atlas(41.2 on QnA, 42.6 RF, 39.1 TW)\n✅ClawEval(77.1)\nPost-trained on top of gemma4 and qwen3.5, Ornith-1.0 employs a novel self-improving training strategy in which reinforcement learning is used to generate not only solution rollouts, but also the task-specific scaffolds that drive those rollouts. By jointly optimizing the scaffold and the resulting solution, the model generate higher-quality solutions in agentic coding.😎\nAll models are released under the MIT license, enabling full commercial and research use.\n📖Tech Blog:", "url": "https://wpnews.pro/news/new-agentic-coding-sota-models", "canonical_source": "https://twitter.com/ornith_/status/2070148887067963854", "published_at": "2026-06-25 15:00:10+00:00", "updated_at": "2026-06-25 15:14:15.299170+00:00", "lang": "en", "topics": ["large-language-models", "ai-agents", "ai-research", "ai-products"], "entities": ["Ornith-1.0", "Gemma 4", "Qwen 3.5", "SWE-Bench", "Terminal-Bench", "NL2Repo", "SWE Atlas", "ClawEval"], "alternates": {"html": "https://wpnews.pro/news/new-agentic-coding-sota-models", "markdown": "https://wpnews.pro/news/new-agentic-coding-sota-models.md", "text": "https://wpnews.pro/news/new-agentic-coding-sota-models.txt", "jsonld": "https://wpnews.pro/news/new-agentic-coding-sota-models.jsonld"}}