[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-glm-5-1-zhipu-744b-moe-40b-mit-swe-bench-pro":3,"topics-all":36,"news-related-7e57ed2c-8271-4a8b-b694-e035328feef7":55},{"id":4,"title":5,"summary":6,"content":6,"original_url":7,"source_id":8,"tags":9,"translations":23,"news_slug":29,"published_at":30,"created_at":31,"modified_at":32,"is_published":33,"publish_type":34,"image_url":13,"view_count":35},"7e57ed2c-8271-4a8b-b694-e035328feef7","GLM-5.1：开源模型的新里程碑","**GLM-5.1：开源模型的新里程碑**\n\n4月7日，智谱AI发布了GLM-5.1，这个7440亿参数的混合专家模型不仅打破了技术壁垒，更在SWE-Bench Pro基准测试中超越了GPT-5.4和Claude Opus 4.6，标志着开源模型正式跻身第一梯队。\n\n**技术突破**：GLM-5.1采用了先进的混合专家架构，每前向传播激活400亿参数，拥有200K的上下文窗口。最令人震撼的是，它完全基于MIT许可证发布，意味着开发者可以不受限制地使用、修改和再分发，甚至用于商业用途。\n\n**性能对比**：在SWE-Bench Pro这一衡量真实世界软件工程能力的基准测试中，GLM-5.1的表现令人瞩目。这项测试要求模型完成专家级别的编程任务，从调试复杂bug到编写大型应用程序。GLM-5.1的领先证明了开源模型在实用性上已经达到了闭源模型的同等水平。\n\n**行业意义**：GLM-5.1的发布意义远超技术本身。它表明开源模型不再是“追赶者”，而是成为了行业的引领者。MIT许可证的采用（相比Apache 2.0更宽松）也体现了智谱AI对开源社区的信任。对于企业而言，这意味着他们现在可以通过自托管的方式运行最强大的编程模型，而无需担心API费用或数据安全。\n\n**未来展望**：随着GLM-5.1这样的模型出现，企业重新评估了构建vs购买的决策天平。当顶级模型可以通过本地部署获得时，传统API服务的价值主张正在被重塑。这可能会加速AI民主化的进程，让更多组织能够利用最前沿的AI技术而不受成本限制。\n\nGLM-5.1的成功不仅是智谱AI的胜利，更是整个开源AI社区的里程碑。它证明了开放协作可以创造出世界级的AI系统，并将在未来几年内继续推动AI技术的民主化进程。","https:\u002F\u002Fzhipu.ai\u002Fblog\u002Fglm-5-1-milestone-open-source","1eab5c4a-0c8e-49a4-8ac8-0f84a2a3c3a4",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"a8002d98-9df1-4ab9-94d4-a7625af634c4","china-ai",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":18,"name":19,"slug":19,"description":13,"color":13},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",{"id":21,"name":22,"slug":22,"description":13,"color":13},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"6ac2e16f-ab93-4477-9c83-b7a47a7def60","en","GLM-5.1: A New Milestone for Open-Source Models","**GLM-5.1: A New Milestone for Open-Source Models**\n\nOn April 7, Zhipu AI released GLM-5.1, a 744B-parameter Mixture of Experts model that not only breaks technical barriers but surpasses GPT-5.4 and Claude Opus 4.6 on the SWE-Bench Pro benchmark — marking open-source models officially joining the top tier.\n\n**Technical Breakthrough:** GLM-5.1 uses an advanced MoE architecture, activates 40B parameters per forward pass, and has a 200K context window. Most strikingly, it is released entirely under the MIT license, meaning developers can use, modify, and redistribute it without restriction — even for commercial use.\n\n**Performance Comparison:** On SWE-Bench Pro — which measures real-world software engineering ability — GLM-5.1's performance is remarkable. The benchmark requires models to complete expert-level programming tasks, from debugging complex bugs to writing large applications. GLM-5.1's lead proves open-source models have reached the same level of practical capability as closed-source models.\n\n**Industry Significance:** GLM-5.1's release is far more than technical. It shows open-source models are no longer \"catch-up players\" but industry leaders. The adoption of the MIT license (more permissive than Apache 2.0) also reflects Zhipu AI's trust in the open-source community. For enterprises, this means they can now self-host the most powerful coding models without worrying about API costs or data security.\n\n**Future Outlook:** With models like GLM-5.1, enterprises are recalibrating the build-vs-buy scale. When top-tier models can be deployed locally, the value proposition of traditional API services is being reshaped. This may accelerate AI democratization, letting more organizations leverage cutting-edge AI without cost barriers. GLM-5.1's success is not just a win for Zhipu AI — it's a milestone for the entire open-source AI community, proving that open collaboration can produce world-class AI systems and will continue to drive AI democratization for years to come.","glm-5-1-zhipu-744b-moe-40b-mit-swe-bench-pro","2026-04-20T03:01:00Z","2026-04-20T07:06:06.648545Z","2026-08-19T02:08:40.142862Z",true,"agent",138,[37,46],{"slug":38,"tag_slug":38,"title_zh":39,"title_en":40,"intro_zh":41,"intro_en":42,"id":43,"is_active":33,"created_at":44,"modified_at":45},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":47,"tag_slug":47,"title_zh":48,"title_en":49,"intro_zh":50,"intro_en":51,"id":52,"is_active":33,"created_at":53,"modified_at":54},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":56},[57,62,67,72,77,82],{"id":58,"title":59,"news_slug":60,"published_at":61},"2b37a19b-1dde-4238-bef5-39b1d19157f1","OpenBMB 开源 MiniCPM5-2B:2B 端侧模型平均分超对比集 4B 级","openbmb-minicpm5-2b-on-device","2026-09-07T17:02:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"453ce9a1-5d55-4981-b44d-c261b8051724","GLM-5.3 753B 权重上架 HuggingFace,智谱兑现两周开源承诺","glm-5-3-weights-huggingface-release","2026-08-28T15:15:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"f6e4aab0-7693-4c2c-bb66-c1641fc2cc3e","Ox Alpha 谜底揭晓:智谱 GLM-5.3-Flash,MIT 开源 320B MoE","ox-alpha-glm-5-3-flash-reveal","2026-08-27T13:30:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"804ab59a-a8d6-4b61-bf74-8f6f2bdae83c","智谱把 Flash 做成一件正经事:一次说清 GLM-5.3-Flash 的架构和 benchmark 真相","glm-5-3-flash-hybrid-attention-architecture","2026-08-27T08:00:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"b0183d10-bcfd-44ed-a178-a2c813f10b69","国家超算互联网AI社区上线Kimi K3:2.8万亿参数MoE一键调用,开源大模型有了国产算力底座","kimi-k3-scnet-platform-launch","2026-07-28T09:30:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"3d8b9b1a-e038-466f-9b6b-304f911e35a7","Kimi K3 开源三件套 MoonEP\u002FFlashKDA\u002FAgentEnv:Moonshot 把 2.8T MoE 训练栈完整交底","kimi-k3-moonep-flashkda-agentenv","2026-07-28T04:30:00+00:00"]