{"slug": "freetoken-efficient-edge-native-moe-serving-with-bandwidth-adaptive-execution", "title": "FreeToken: Efficient Edge-Native Moe Serving with Bandwidth-Adaptive Execution", "summary": "Researchers introduced FreeToken, an edge-native Mixture-of-Experts (MoE) serving system that enables frontier-scale AI models to run on personal hardware, supporting over 20 MoE models and real coding and tool-using agents across devices from an 8GB laptop GPU to a single workstation GPU. The system expands practical serving from a 35B model on a laptop to a 284B model on a gaming desktop and the 753B GLM-5.2 on a single workstation GPU, turning open weights into deployable local software. FreeToken co-designs the full serving stack with bandwidth-adaptive execution to handle heterogeneous edge resources and continuously changing agent workloads.", "body_md": "# Computer Science > Distributed, Parallel, and Cluster Computing\n\n[Submitted on 17 Aug 2026]\n\n# Title:FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution\n\n[View PDF](/pdf/2608.16157)\n\n[HTML (experimental)](https://arxiv.org/html/2608.16157v1)\n\nAbstract:Frontier open-weight models are increasingly available, but serving them still largely assumes datacenter infrastructure. We present FreeToken, an edge-native MoE serving system that treats a personal machine not as a small GPU, but as a unified, elastic inference platform. FreeToken co-designs the full serving stack, including model layout and loading, expert residency, CPU--GPU execution, agentic state reuse, and runtime memory management, around two realities of local AI: agent workloads continuously change their execution pattern, and edge hardware exposes heterogeneous resources whose balance differs from machine to machine. Rather than committing to a fixed offloading strategy, FreeToken continuously maps computation and model state onto the resources actually available. FreeToken supports more than 20 MoE models and real coding and tool-using agents across hardware ranging from an 8GB laptop GPU to a single workstation GPU. More importantly, it changes what these machines can practically serve, from a 35B model on a laptop to a 284B model on a gaming desktop and the 753B GLM-5.2 on a single workstation GPU. FreeToken turns open weights into deployable local software, making the machines users already own a practical platform for frontier-scale intelligence. We release the system at[this http URL].\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/freetoken-efficient-edge-native-moe-serving-with-bandwidth-adaptive-execution", "canonical_source": "https://arxiv.org/abs/2608.16157", "published_at": "2026-08-21 23:49:16+00:00", "updated_at": "2026-08-22 00:13:24.985573+00:00", "lang": "en", "topics": ["machine-learning", "ai-infrastructure"], "entities": ["FreeToken", "GLM-5.2"], "alternates": {"html": "https://wpnews.pro/news/freetoken-efficient-edge-native-moe-serving-with-bandwidth-adaptive-execution", "markdown": "https://wpnews.pro/news/freetoken-efficient-edge-native-moe-serving-with-bandwidth-adaptive-execution.md", "text": "https://wpnews.pro/news/freetoken-efficient-edge-native-moe-serving-with-bandwidth-adaptive-execution.txt", "jsonld": "https://wpnews.pro/news/freetoken-efficient-edge-native-moe-serving-with-bandwidth-adaptive-execution.jsonld"}}