{"slug": "rufus-air-an-open-llm-post-training-recipe", "title": "Rufus-Air: An Open LLM Post-Training Recipe", "summary": "Rufus-Air is an open, reproducible post-training recipe built on GLM-4.5-Air-Base (106B-A12B) that runs as a serial pipeline of eight stages: SFT, Reasoning RL, Coding RL, Instruction-Following RL, General Agent, Coding Agent, Search Agent, and RLHF. The recipe documents its data, reward design, and infrastructure.", "body_md": "Rufus-Air is an open and reproducible post-training recipe on GLM-4.5-Air-Base (106B-A12B), organized as a serial pipeline of eight stages: SFT, Reasoning RL, Coding RL, Instruction-Following RL, General Agent, Coding Agent, Search Agent, and RLHF. We document the data, reward design, infrastructure", "url": "https://wpnews.pro/news/rufus-air-an-open-llm-post-training-recipe", "canonical_source": "https://aiflash.com/news/125985/", "published_at": "2026-09-25 06:30:09+00:00", "updated_at": "2026-09-25 06:59:14.508049+00:00", "lang": "en", "topics": ["large-language-models", "ai-research", "machine-learning", "ai-agents", "artificial-intelligence"], "entities": ["Rufus-Air", "GLM-4.5-Air-Base"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/rufus-air-an-open-llm-post-training-recipe", "markdown": "https://wpnews.pro/news/rufus-air-an-open-llm-post-training-recipe.md", "text": "https://wpnews.pro/news/rufus-air-an-open-llm-post-training-recipe.txt", "jsonld": "https://wpnews.pro/news/rufus-air-an-open-llm-post-training-recipe.jsonld"}}