{"slug": "loop-the-loopies", "title": "Loop the Loopies", "summary": "Researchers have introduced the Loopie series, two Mixture-of-Experts models (20B parameters with 2B active and 6B with 0.6B active) that outperform vanilla Transformer baselines trained with the same compute budget, addressing the long-standing challenge that increasing parameter count outperforms looping. A novel post-training method gives Loopie frontier-level reasoning abilities.", "body_md": "# Computer Science > Computation and Language\n\n[Submitted on 17 Jul 2026 (\n\n[v1](https://arxiv.org/abs/2607.16051v1)), last revised 20 Jul 2026 (this version, v2)]# Title:Loop the Loopies!\n\n[View PDF](/pdf/2607.16051)\n\n[HTML (experimental)](https://arxiv.org/html/2607.16051v2)\n\nAbstract:We present the Loopie series, consisting of two Mixture-of-Experts (MoE) models: a 20B-parameter model with 2B active parameters and a 6B-parameter model with 0.6B active parameters. Looped Transformers have long faced a challenge: given an N times increase in pre-training compute, increasing the parameter count by a factor of N usually outperforms looping a model N times. Loopie addresses this challenge. Extensive ablation studies, including comparisons with a vanilla 30B-A3B model, show that Loopie substantially outperforms vanilla Transformer baselines trained with the same compute budget. With a novel post-training method, Loopie develops strong reasoning abilities and achieves frontier-level reasoning performance.\n\n## Submission history\n\nFrom: Zitian Gao [[view email](/show-email/52340bf5/2607.16051)]\n\n**Fri, 17 Jul 2026 15:28:43 UTC (829 KB)**\n\n[[v1]](/abs/2607.16051v1)**[v2]** Mon, 20 Jul 2026 15:59:50 UTC (835 KB)\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/loop-the-loopies", "canonical_source": "https://arxiv.org/abs/2607.16051", "published_at": "2026-07-21 13:56:44+00:00", "updated_at": "2026-07-21 14:13:09.213765+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research"], "entities": ["Loopie", "Mixture-of-Experts", "Transformer"], "alternates": {"html": "https://wpnews.pro/news/loop-the-loopies", "markdown": "https://wpnews.pro/news/loop-the-loopies.md", "text": "https://wpnews.pro/news/loop-the-loopies.txt", "jsonld": "https://wpnews.pro/news/loop-the-loopies.jsonld"}}