{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-09-28T06:03:00.468Z","headline":"开源动态第 24 期：Motif-3、GLM-5.3、Hy4-preview 与开源模型许可证趋势","description":"本期开源模型汇总指出，2026 年开源模型竞争加剧带来许可证分化：Google 和 Meta 转向 Apache 2.0，而智谱 GLM-5.3 从 MIT 改为自定义许可证，要求年收入超 100 亿美元的推理或微调服务商通过 Z.AI 安全审查，且未定义关联方带来不确定性。","url":"https://www.aioga.com/news/cmtssoxb003vvrokaz4fxgjpl/","mainEntityOfPage":"https://www.aioga.com/news/cmtssoxb003vvrokaz4fxgjpl/","datePublished":"2026-09-08T14:15:25.000Z","dateModified":"2026-09-08T14:15:25.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3","https://aihot.news/items/cmtssoxb003vvrokaz4fxgjpl"],"canonicalUrl":"https://www.aioga.com/news/cmtssoxb003vvrokaz4fxgjpl/","directAnswer":{"@type":"Answer","text":"本期汇总称，2026年开源模型竞争加剧并伴随许可证分化：Google和Meta采用Apache 2.0，智谱GLM-5.3由MIT改为自定义许可证，Motif-3采用MIT许可证。","url":"https://www.aioga.com/news/cmtssoxb003vvrokaz4fxgjpl/","dateCreated":"2026-09-08T14:15:25.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"interconnects.ai source article","url":"https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3","datePublished":"2026-09-08T14:15:25.000Z","provider":{"@type":"Organization","name":"interconnects.ai","url":"https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.news/items/cmtssoxb003vvrokaz4fxgjpl","datePublished":"2026-09-08T14:15:25.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.news/items/cmtssoxb003vvrokaz4fxgjpl"}}],"aggregationSource":"Nathan Lambert：Interconnects（RSS）","originalPublisher":{"name":"interconnects.ai","url":"https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3"},"geoDeepAnswer":null,"article":{"id":"cmtssoxb003vvrokaz4fxgjpl","slug":"cmtssoxb003vvrokaz4fxgjpl","url":"https://www.aioga.com/news/cmtssoxb003vvrokaz4fxgjpl/","title":"开源动态第 24 期：Motif-3、GLM-5.3、Hy4-preview 与开源模型许可证趋势","title_en":"","summary":"本期开源模型汇总指出，2026 年开源模型竞争加剧带来许可证分化：Google 和 Meta 转向 Apache 2.0，而智谱 GLM-5.3 从 MIT 改为自定义许可证，要求年收入超 100 亿美元的推理或微调服务商通过 Z.AI 安全审查，且未定义关联方带来不确定性。","source":"Nathan Lambert：Interconnects（RSS）","sourceUrl":"https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3","aiHotUrl":"https://aihot.news/items/cmtssoxb003vvrokaz4fxgjpl","publishedAt":"2026-09-08T14:15:25.000Z","category":"行业动态","score":58,"selected":false,"articleBody":["Avid Artifacts readers know that we have been covering not only models but also their licenses for quite some time. There was a period when custom licenses were all the rage, for example the custom Qwen2.5 72B-Instruct license：https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE or the Llama licenses：https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct/blob/main/LICENSE . DeepSeek had a custom license for DeepSeek V3：https://huggingface.co/deepseek-ai/DeepSeek-V3/blob/main/LICENSE-MODEL before R1 changed it to MIT：https://huggingface.co/deepseek-ai/DeepSeek-R1/blob/main/LICENSE , which has resulted in many (Chinese) model makers adopting MIT or Apache 2.0 licenses in 2025.","In 2026, open models are more competitive than ever, which has led to two interesting developments: Western model makers adopt open licenses, with both Google：https://huggingface.co/google/gemma-4-31B-it and Meta：https://huggingface.co/meta-models/Muse-Glimmer-30B switching to Apache 2.0. Chinese model makers at the frontier, however, are becoming more restrictive: Kimi K3：https://huggingface.co/moonshotai/Kimi-K3 comes with a license which requires commercial agreements for those who run inference or fine-tuning services, and MiniMax M3：https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE requires agreements above a revenue threshold and has prohibited use cases.","The newest addition is Zhipu’s GLM-5.3：https://huggingface.co/zai-org/GLM-5.3 , which switched from MIT (GLM-5.2 and earlier) to a custom license with the following clause for inference and fine-tuning providers:","If the Licensee or any of its affiliates operates a Model as a Service business, and the aggregate revenue of the Licensee and its affiliates exceeds 10 billion US dollars (or the equivalent in other currencies) in total over any consecutive 12 months, the Licensee must pass Z.AI’s security review before using the Software or its derivative works for any commercial purpose. The scope and method of the security review shall be reasonably determined by Z.AI.","While the 10 billion US dollar threshold is very high compared to other licenses of this kind, “affiliates” is not defined in the license, which adds uncertainty and creates barriers to adoption. Furthermore, the license is provided in both English and Chinese, with the Chinese text using “关联方” for affiliated parties, which does have a definition in Chinese law：https://kjs.mof.gov.cn/zt/kjzzss/kuaijizhunzeshishi/200806/t20080618_46245.htm .","We are by no means legal experts and there are obvious reasons why those licenses are created. However, we want to highlight the issues that come with creating such licenses, especially in a world with a lot of valid open and closed alternatives.","Share ：https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3?utm_source=substack&utm_medium=email&utm_content=share&action=share","Motif-3 ：https://huggingface.co/Motif-Technologies/Motif-3 by Motif-Technologies ：https://huggingface.co/Motif-Technologies : Motif is one of the few hidden gems out there, showcasing innovation in their model training with very limited resources compared to others. Motif-3 comes with an MIT license and impressive scores for its size. Given the trajectory of model releases from Motif 2.6B, which we covered in 2025 ：https://artifactshub.ai/Motif-Technologies/motif-2-6b and Motif-2-12.7B ：https://artifactshub.ai/Motif-Technologies/motif-2-12-7b-instruct , the improvements are impressive.","dots3-note-prev ：https://huggingface.co/dots-studio/dots3-note-prev by dots-studio ：https://huggingface.co/dots-studio : RedNote/Xiaohongshu, the Chinese Instagram, is also getting more serious about model training, although they aren’t exactly a newcomer ：https://artifactshub.ai/rednote-hilab/dots-llm1-inst , having released models as early as 2025. dots3 was also able to win the IMO 2026 with a perfect score using an internal harness. We expect more from them in the near future.","Qwen3.8-Flash-Next ：https://huggingface.co/Qwen/Qwen3.8-Flash-Next by Qwen ：https://huggingface.co/Qwen : A preview of the next version of Qwen models in terms of architecture: 125B-A6B with 51B n-gram embeddings. It uses GDN and Qwen Sparse Attention. Similar to Qwen3-Next-80B-A3B-Instruct ：https://artifactshub.ai/Qwen/qwen3-next-80b-a3b-instruct , we expect similar architectures to become more popular and the ecosystem to fix integrations by the time Qwen4 drops.","GLM-5.3-Flash ：https://huggingface.co/zai-org/GLM-5.3-Flash by zai-org ：https://huggingface.co/zai-org : This release perfected the version of the Chinese model playbook we’ve written about in 2025 ：https://www.interconnects.ai/p/latest-open-artifacts-16-whos-building : The model got released as a free-to-use “stealth model” under the name “Ox-Alpha” on OpenRouter and OpenCode, which got people excited to try it out in the first place. They then speculated about its creator and size, alleging it is a >1T model from Cursor/xAI, Gemini or a new pre-train from open source labs. Because the model is relatively performant, people kept speculating for days about its creator, thus building up hype. It also dampens the accusations of benchmaxxing which accompany every (open) model release.","Hy4-preview ：https://huggingface.co/tencent/Hy4-preview by tencent ：https://huggingface.co/tencent : Tencent is becoming a serious player in the open model space, increasing the size of their flagship model while spinning the post-training flywheel. The result, Hy4-preview, is a competent model which currently has an issue with overthinking. However, if the trajectory from Hy3-preview ：https://artifactshub.ai/tencent/hy3-preview to Hy3 ：https://artifactshub.ai/tencent/hy3 is any indication, the final model might be a legit shot at the front ranks of open models.","View more details on all the models in this issue at our Artifacts Hub：https://artifactshub.ai/?picks=0 .","Visit artifactshub.ai ：https://artifactshub.ai/?picks=0","NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 ：https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 by nvidia ：https://huggingface.co/nvidia : An update to Nemotron, which comes with performance — but especially speed improvements — across the board.","Ling-3.0-flash ：https://huggingface.co/inclusionAI/Ling-3.0-flash by inclusionAI ：https://huggingface.co/inclusionAI : Ant Ling is a frequent guest at the Artifacts Log; they are now on their third iteration of models, adopting a hybrid design (KDA + Gated MLA), similar to others. They also release a small 7.9B-A1.3B ：https://huggingface.co/inclusionAI/Ling-3.0-tiny version.","Qwen3.8-2.4T-A95B ：https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B by Qwen ：https://huggingface.co/Qwen : In a rather surprising turn of events, Alibaba started to openly release their biggest versions of Qwen as well. However, it comes with a custom license and its performance is behind other models of its size."],"articleImages":[],"mediaStatus":"none","articleBodyZh":["Avid Artifacts 的读者都知道，我们不仅长期关注模型本身，也关注它们的许可证。有一段时间，定制许可证非常流行，例如定制的 Qwen2.5 72B-Instruct 许可证：https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE 或 Llama 许可证：https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct/blob/main/LICENSE。DeepSeek 曾为 DeepSeek V3 制定了定制许可证：https://huggingface.co/deepseek-ai/DeepSeek-V3/blob/main/LICENSE-MODEL，后来 R1 将其改为 MIT：https://huggingface.co/deepseek-ai/DeepSeek-R1/blob/main/LICENSE，这导致许多中国模型开发者在 2025 年采用 MIT 或 Apache 2.0 许可证。","到 2026 年，开放模型的竞争比以往任何时候都更激烈，这带来了两个有趣的发展：西方模型开发者采用开放许可证，例如 Google：https://huggingface.co/google/gemma-4-31B-it 和 Meta：https://huggingface.co/meta-models/Muse-Glimmer-30B 都转向 Apache 2.0。而中国前沿的模型开发者则变得更加严格：Kimi K3：https://huggingface.co/moonshotai/Kimi-K3 配备了要求商业使用者签署协议的许可证，MiniMax M3：https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE 则要求收入超过阈值的用户签订协议，并禁止某些使用场景。","最新加入的是智谱的 GLM-5.3：https://huggingface.co/zai-org/GLM-5.3，它从 MIT（GLM-5.2 及以前版本）转为定制许可证，其中对推理和微调服务提供者包含以下条款：","如果被许可方或其任何关联公司经营模型即服务（Model as a Service）业务，并且被许可方及其关联公司在任意连续 12 个月内的总收入超过 100 亿美元（或其他等值货币），则被许可方在将软件或其衍生作品用于任何商业目的之前，必须通过 Z.AI 的安全审查。安全审查的范围和方法由 Z.AI 合理确定。","与其他同类许可相比，100亿美元的门槛非常高，但许可中并未定义“关联方”，这增加了不确定性并造成采用障碍。此外，该许可提供英文和中文版本，中文文本中使用“关联方”来表示 affiliated parties，而在中国法律中确实有定义：https://kjs.mof.gov.cn/zt/kjzzss/kuaijizhunzeshishi/200806/t20080618_46245.htm。","我们绝不是法律专家，而且这些许可的制定有明显理由。然而，我们希望强调创建此类许可所带来的问题，尤其是在存在许多有效开源和闭源替代方案的情况下。","分享：https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3?utm_source=substack&utm_medium=email&utm_content=share&action=share","Motif-3：https://huggingface.co/Motif-Technologies/Motif-3 由 Motif-Technologies 提供：https://huggingface.co/Motif-Technologies：Motif 是少数几个隐藏的瑰宝之一，在模型训练上展示了创新性，即使资源远少于其他人。Motif-3 采用 MIT 许可，并且以其体量取得了令人印象深刻的成绩。根据我们在 2025 年覆盖的 Motif 2.6B：https://artifactshub.ai/Motif-Technologies/motif-2-6b 和 Motif-2-12.7B：https://artifactshub.ai/Motif-Technologies/motif-2-12-7b-instruct 发布的模型轨迹来看，其改进令人印象深刻。","dots3-note-prev：https://huggingface.co/dots-studio/dots3-note-prev 由 dots-studio 提供：https://huggingface.co/dots-studio：RedNote/Xiaohongshu，中国的 Instagram，也在更认真地进行模型训练，尽管他们并非新人：https://artifactshub.ai/rednote-hilab/dots-llm1-inst，并且早在 2025 年就已经发布了模型。dots3 还能够凭借内部工具在 IMO 2026 中获得满分。我们期待他们在不久的将来带来更多成果。","Qwen3.8-Flash-Next：https://huggingface.co/Qwen/Qwen3.8-Flash-Next 由 Qwen 提供：https://huggingface.co/Qwen：对 Qwen 模型下一版本的架构预览：125B-A6B，带有 51B n-gram 嵌入。它使用 GDN 和 Qwen 稀疏注意力。类似于 Qwen3-Next-80B-A3B-Instruct：https://artifactshub.ai/Qwen/qwen3-next-80b-a3b-instruct，我们预计相似的架构将会更受欢迎，并且在 Qwen4 发布时生态系统的集成问题将会得到解决。","GLM-5.3-Flash：https://huggingface.co/zai-org/GLM-5.3-Flash 由 zai-org 提供：https://huggingface.co/zai-org：此版本完善了我们在 2025 年撰写的中文模型手册：https://www.interconnects.ai/p/latest-open-artifacts-16-whos-building：该模型作为免费的“隐形模型”以“Ox-Alpha”的名义在 OpenRouter 和 OpenCode 上发布，这最初激发了人们尝试它的兴趣。然后，人们对其创作者和规模进行推测，声称它是 Cursor/xAI、Gemini 或开源实验室的新预训练模型，规模超过 1T。由于该模型性能相对出色，人们连续数天猜测其创作者，从而炒热了话题。这也减轻了每次（开放）模型发布伴随的极限测试指控。","Hy4-preview：https://huggingface.co/tencent/Hy4-preview 由腾讯提供：https://huggingface.co/tencent：腾讯正在成为开放模型领域的一个重要参与者，增加其旗舰模型的规模，同时推动后训练飞轮。结果，Hy4-preview 是一个能力不错的模型，目前存在过度思考的问题。不过，如果从 Hy3-preview：https://artifactshub.ai/tencent/hy3-preview 到 Hy3：https://artifactshub.ai/tencent/hy3 的发展轨迹可以参考，最终模型可能会有机会跻身开放模型的前列。","在我们的 Artifacts Hub 查看本期所有模型的更多详细信息：https://artifactshub.ai/?picks=0。","访问 artifactshub.ai：https://artifactshub.ai/?picks=0","NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16：https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 由 nvidia 提供：https://huggingface.co/nvidia：Nemotron 的更新版本，在性能方面有提升——特别是整体速度的改进。","Ling-3.0-flash：https://huggingface.co/inclusionAI/Ling-3.0-flash 由 inclusionAI 提供：https://huggingface.co/inclusionAI：Ant Ling 经常出现在 Artifacts Log 上；他们现在已经是模型的第三次迭代，采用混合设计（KDA + Gated MLA），类似其他模型。他们还发布了一个小型 7.9B-A1.3B 版本：https://huggingface.co/inclusionAI/Ling-3.0-tiny。","Qwen3.8-2.4T-A95B：https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B 由 Qwen 提供：https://huggingface.co/Qwen：在一个相当令人惊讶的事件中，阿里巴巴也开始公开发布他们最大的 Qwen 版本。然而，它附带自定义许可，并且其性能落后于同规模的其他模型。"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"本期汇总称，2026年开源模型竞争加剧并伴随许可证分化：Google和Meta采用Apache 2.0，智谱GLM-5.3由MIT改为自定义许可证，Motif-3采用MIT许可证。","background":"来源回顾称，DeepSeek V3曾使用自定义许可证，R1改为MIT后，2025年有多家中国模型厂商采用MIT或Apache 2.0。Kimi K3、MiniMax M3和GLM-5.3则包含商业协议、收入门槛或限制性条款。","viewpoint":"Aioga 判断：GLM-5.3的相关条款规定，许可证持有人及其关联方经营模型服务且连续12个月合计收入超过100亿美元时，商业使用软件或衍生作品前需通过Z.AI安全审查；“关联方”未在许可证中定义。","implications":"可能影响：推理或微调服务商需要评估自身及关联方是否达到100亿美元收入门槛，并核对安全审查要求；条款未定义“关联方”，可能增加合规判断成本，但不代表所有使用者都会触发审查。","nextStep":"后续观察：应继续核对GLM-5.3许可证的中英文表述、关联方适用范围及安全审查细则，并与MIT、Apache 2.0及其他自定义许可证的实际使用条件比较。","evidenceRefs":["title","summary","articleBody","source"],"status":"published","aiGenerated":true,"autoApproved":true,"generatedBy":"aioga-editorial:gpt-5.6-sol","reviewedBy":"aioga-editorial-review:gpt-5.6-sol","generatedAt":"2026-09-08T15:44:52.458Z","sourceHash":"6ca5d3d57445fac5","review":{"approved":true,"groundedness":96,"clarity":92,"duplicationRisk":22,"blockingIssues":[],"notes":["“Google和Meta采用Apache 2.0”可改为“转向Apache 2.0”或“切换至Apache 2.0”，以更准确保留来源中的变化语义。","“Aioga判断”属于编辑观点标注；如无必要，可改为“分析认为”或直接保留为观点栏内容。","“后续观察”中的核对与比较属于合理的编辑建议，不构成来源事实断言。"]},"validation":{"passed":true,"mode":"ai-auto","revisions":1,"checks":["schema","length","source-attribution","editorial-labels","inference-boundary","low-source-overlap","no-html","independent-ai-review"]}},"tags":["行业动态","Nathan Lambert：Interconnects（RSS）"],"translations":{"zh-CN":{"title":"开源动态第 24 期：Motif-3、GLM-5.3、Hy4-preview 与开源模型许可证趋势","summary":"本期开源模型汇总指出，2026 年开源模型竞争加剧带来许可证分化：Google 和 Meta 转向 Apache 2.0，而智谱 GLM-5.3 从 MIT 改为自定义许可证，要求年收入超 100 亿美元的推理或微调服务商通过 Z.AI 安全审查，且未定义关联方带来不确定性。","category":"行业动态","source":"interconnects.ai","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"开源动态第 24 期：Motif-3、GLM-5.3、Hy4-preview 与开源模型许可证趋势 - Aioga AI资讯","description":"本期开源模型汇总指出，2026 年开源模型竞争加剧带来许可证分化：Google 和 Meta 转向 Apache 2.0，而智谱 GLM-5.3 从 MIT 改为自定义许可证，要求年收入超 100 亿美元的推理或微调服务商通过 Z.AI 安全审查，且未定义关联方带来不确定性。","url":"https://www.aioga.com/news/cmtssoxb003vvrokaz4fxgjpl/","articleBody":["Avid Artifacts 的读者都知道，我们不仅长期关注模型本身，也关注它们的许可证。有一段时间，定制许可证非常流行，例如定制的 Qwen2.5 72B-Instruct 许可证：https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE 或 Llama 许可证：https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct/blob/main/LICENSE。DeepSeek 曾为 DeepSeek V3 制定了定制许可证：https://huggingface.co/deepseek-ai/DeepSeek-V3/blob/main/LICENSE-MODEL，后来 R1 将其改为 MIT：https://huggingface.co/deepseek-ai/DeepSeek-R1/blob/main/LICENSE，这导致许多中国模型开发者在 2025 年采用 MIT 或 Apache 2.0 许可证。","到 2026 年，开放模型的竞争比以往任何时候都更激烈，这带来了两个有趣的发展：西方模型开发者采用开放许可证，例如 Google：https://huggingface.co/google/gemma-4-31B-it 和 Meta：https://huggingface.co/meta-models/Muse-Glimmer-30B 都转向 Apache 2.0。而中国前沿的模型开发者则变得更加严格：Kimi K3：https://huggingface.co/moonshotai/Kimi-K3 配备了要求商业使用者签署协议的许可证，MiniMax M3：https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE 则要求收入超过阈值的用户签订协议，并禁止某些使用场景。","最新加入的是智谱的 GLM-5.3：https://huggingface.co/zai-org/GLM-5.3，它从 MIT（GLM-5.2 及以前版本）转为定制许可证，其中对推理和微调服务提供者包含以下条款：","如果被许可方或其任何关联公司经营模型即服务（Model as a Service）业务，并且被许可方及其关联公司在任意连续 12 个月内的总收入超过 100 亿美元（或其他等值货币），则被许可方在将软件或其衍生作品用于任何商业目的之前，必须通过 Z.AI 的安全审查。安全审查的范围和方法由 Z.AI 合理确定。","与其他同类许可相比，100亿美元的门槛非常高，但许可中并未定义“关联方”，这增加了不确定性并造成采用障碍。此外，该许可提供英文和中文版本，中文文本中使用“关联方”来表示 affiliated parties，而在中国法律中确实有定义：https://kjs.mof.gov.cn/zt/kjzzss/kuaijizhunzeshishi/200806/t20080618_46245.htm。","我们绝不是法律专家，而且这些许可的制定有明显理由。然而，我们希望强调创建此类许可所带来的问题，尤其是在存在许多有效开源和闭源替代方案的情况下。","分享：https://www.interconnects.ai/p/latest-open-artifacts-24-motif-3?utm_source=substack&utm_medium=email&utm_content=share&action=share","Motif-3：https://huggingface.co/Motif-Technologies/Motif-3 由 Motif-Technologies 提供：https://huggingface.co/Motif-Technologies：Motif 是少数几个隐藏的瑰宝之一，在模型训练上展示了创新性，即使资源远少于其他人。Motif-3 采用 MIT 许可，并且以其体量取得了令人印象深刻的成绩。根据我们在 2025 年覆盖的 Motif 2.6B：https://artifactshub.ai/Motif-Technologies/motif-2-6b 和 Motif-2-12.7B：https://artifactshub.ai/Motif-Technologies/motif-2-12-7b-instruct 发布的模型轨迹来看，其改进令人印象深刻。","dots3-note-prev：https://huggingface.co/dots-studio/dots3-note-prev 由 dots-studio 提供：https://huggingface.co/dots-studio：RedNote/Xiaohongshu，中国的 Instagram，也在更认真地进行模型训练，尽管他们并非新人：https://artifactshub.ai/rednote-hilab/dots-llm1-inst，并且早在 2025 年就已经发布了模型。dots3 还能够凭借内部工具在 IMO 2026 中获得满分。我们期待他们在不久的将来带来更多成果。","Qwen3.8-Flash-Next：https://huggingface.co/Qwen/Qwen3.8-Flash-Next 由 Qwen 提供：https://huggingface.co/Qwen：对 Qwen 模型下一版本的架构预览：125B-A6B，带有 51B n-gram 嵌入。它使用 GDN 和 Qwen 稀疏注意力。类似于 Qwen3-Next-80B-A3B-Instruct：https://artifactshub.ai/Qwen/qwen3-next-80b-a3b-instruct，我们预计相似的架构将会更受欢迎，并且在 Qwen4 发布时生态系统的集成问题将会得到解决。","GLM-5.3-Flash：https://huggingface.co/zai-org/GLM-5.3-Flash 由 zai-org 提供：https://huggingface.co/zai-org：此版本完善了我们在 2025 年撰写的中文模型手册：https://www.interconnects.ai/p/latest-open-artifacts-16-whos-building：该模型作为免费的“隐形模型”以“Ox-Alpha”的名义在 OpenRouter 和 OpenCode 上发布，这最初激发了人们尝试它的兴趣。然后，人们对其创作者和规模进行推测，声称它是 Cursor/xAI、Gemini 或开源实验室的新预训练模型，规模超过 1T。由于该模型性能相对出色，人们连续数天猜测其创作者，从而炒热了话题。这也减轻了每次（开放）模型发布伴随的极限测试指控。","Hy4-preview：https://huggingface.co/tencent/Hy4-preview 由腾讯提供：https://huggingface.co/tencent：腾讯正在成为开放模型领域的一个重要参与者，增加其旗舰模型的规模，同时推动后训练飞轮。结果，Hy4-preview 是一个能力不错的模型，目前存在过度思考的问题。不过，如果从 Hy3-preview：https://artifactshub.ai/tencent/hy3-preview 到 Hy3：https://artifactshub.ai/tencent/hy3 的发展轨迹可以参考，最终模型可能会有机会跻身开放模型的前列。","在我们的 Artifacts Hub 查看本期所有模型的更多详细信息：https://artifactshub.ai/?picks=0。","访问 artifactshub.ai：https://artifactshub.ai/?picks=0","NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16：https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 由 nvidia 提供：https://huggingface.co/nvidia：Nemotron 的更新版本，在性能方面有提升——特别是整体速度的改进。","Ling-3.0-flash：https://huggingface.co/inclusionAI/Ling-3.0-flash 由 inclusionAI 提供：https://huggingface.co/inclusionAI：Ant Ling 经常出现在 Artifacts Log 上；他们现在已经是模型的第三次迭代，采用混合设计（KDA + Gated MLA），类似其他模型。他们还发布了一个小型 7.9B-A1.3B 版本：https://huggingface.co/inclusionAI/Ling-3.0-tiny。","Qwen3.8-2.4T-A95B：https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B 由 Qwen 提供：https://huggingface.co/Qwen：在一个相当令人惊讶的事件中，阿里巴巴也开始公开发布他们最大的 Qwen 版本。然而，它附带自定义许可，并且其性能落后于同规模的其他模型。"]},"en":{"title":"Open Source Updates Issue 24: Motif-3, GLM-5.3, Hy4-preview, and Trends in Open Source Model Licenses","summary":"This issue of the Open Source Model Summary points out that in 2026, intensified competition in open source models will lead to license differentiation: Google and Meta will shift to Apache 2.0, while Zhipu's GLM-5.3 will switch from MIT to a custom license, requiring inference or fine-tuning service providers with annual revenues exceeding $10 billion to pass Z.AI security reviews, and undefining related parties brings uncertainty.","category":"Industry","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source Updates Issue 24: Motif-3, GLM-5.3, Hy4-preview, and Trends in Open Source Model Licenses - Aioga AI News","description":"This issue of the Open Source Model Summary points out that in 2026, intensified competition in open source models will lead to license differentiation: Google and Meta will shift...","url":"https://www.aioga.com/en/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:23:51.628Z"},"ja":{"title":"Open Source News 第24号:Motip-3、GLM-5.3、Hy4プレビュー、そしてオープンソースモデルライセンスの動向","summary":"本号のオープンソースモデル概要では、2026年にはオープンソースモデルの競争激化がライセンス差別化につながると指摘しています。GoogleとMetaはARPACHE 2.0に切り替え、ZhipuのGLM-5.3はMITからカスタムライセンスに移行するため、年間収益が100億ドルを超えるサービスプロバイダーはセキュリティ審査を通過するために推論または微調整が必要となり、不明な関係者からの不確実性 Z.AI 不確実性が不確実性をもたらします。","category":"業界動向","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News 第24号:Motip-3、GLM-5.3、Hy4プレビュー、そしてオープンソースモデルライセンスの動向 - Aioga AIニュース","description":"本号のオープンソースモデル概要では、2026年にはオープンソースモデルの競争激化がライセンス差別化につながると指摘しています。GoogleとMetaはARPACHE 2.0に切り替え、ZhipuのGLM-5.3はMITからカスタムライセンスに移行するため、年間収益が100億ドルを超えるサービスプロバイダーはセキュリティ審査を通過するために推論または微調整が必...","url":"https://www.aioga.com/ja/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:23:52.491Z"},"ko":{"title":"오픈 소스 뉴스 24호: Motip-3, GLM-5.3, Hy4-preview, 그리고 오픈 소스 모델 라이선스의 동향","summary":"이번 오픈 소스 모델 요약 호에서는 2026년에 오픈 소스 모델 경쟁이 격화되면서 라이선스 차별화가 발생할 것임을 지적합니다: 구글과 메타는 ARPACHE 2.0으로 전환하고, Zhipu의 GLM-5.3은 MIT에서 맞춤형 라이선스로 전환하며, 연간 매출 100억 달러를 초과하는 서비스 제공자는 보안 검토를 통과해야 추론 또는 미세 조정이 필요하며, 불특정 당사자의 불확실성 Z.AI 불확실성을 초래할 것입니다.","category":"업계 동향","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"오픈 소스 뉴스 24호: Motip-3, GLM-5.3, Hy4-preview, 그리고 오픈 소스 모델 라이선스의 동향 - Aioga AI 뉴스","description":"이번 오픈 소스 모델 요약 호에서는 2026년에 오픈 소스 모델 경쟁이 격화되면서 라이선스 차별화가 발생할 것임을 지적합니다: 구글과 메타는 ARPACHE 2.0으로 전환하고, Zhipu의 GLM-5.3은 MIT에서 맞춤형 라이선스로 전환하며, 연간 매출 100억 달러를 초과하는 서비스 제공자는 보안 검토를 통과해야 추론...","url":"https://www.aioga.com/ko/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:02.056Z"},"es":{"title":"Open Source News Número 24: Motip-3, GLM-5.3, Hy4-preview y tendencias en licencias de modelos de código abierto","summary":"Esta edición del Resumen del Modelo de Código Abierto señala que en 2026, la intensificación de la competencia en modelos de código abierto llevará a la diferenciación de licencias: Google y Meta cambiarán a ARPACHE 2.0, mientras que el GLM-5.3 de Zhipu pasará de MIT a una licencia personalizada, exigiendo que proveedores de servicios de inferencia o ajuste fino con ingresos anuales superiores a 10.000 millones de dólares superen Z.AI revisión de seguridad, y la incertidumbre de partes no especificadas generará incertidumbre.","category":"Industria","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News Número 24: Motip-3, GLM-5.3, Hy4-preview y tendencias en licencias de modelos de código abierto - Aioga Noticias de IA","description":"Esta edición del Resumen del Modelo de Código Abierto señala que en 2026, la intensificación de la competencia en modelos de código abierto llevará a la diferenciación de licencias...","url":"https://www.aioga.com/es/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:02.246Z"},"fr":{"title":"Open Source News Numéro 24 : Motip-3, GLM-5.3, Hy4-preview et tendances en licence de modèles open source","summary":"Ce numéro du résumé du modèle open source souligne qu’en 2026, une concurrence accrue dans les modèles open source entraînera une différenciation des licences : Google et Meta passeront à ARPACHE 2.0, tandis que le GLM-5.3 de Zhipu passera du MIT à une licence personnalisée, exigeant que les prestataires de services d’inférence ou d’ajustement finement dépassent 10 milliards de dollars pour réussir Z.AI revue de sécurité, et l’incertitude venant de parties non spécifiées apportera de l’incertitude.","category":"Industrie","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News Numéro 24 : Motip-3, GLM-5.3, Hy4-preview et tendances en licence de modèles open source - Aioga Actualités IA","description":"Ce numéro du résumé du modèle open source souligne qu’en 2026, une concurrence accrue dans les modèles open source entraînera une différenciation des licences : Google et Meta pass...","url":"https://www.aioga.com/fr/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:11.502Z"},"de":{"title":"Open Source News Ausgabe 24: Motip-3, GLM-5.3, Hy4-Vorschau und Trends in der Open-Source-Modelllizenzierung","summary":"Diese Ausgabe der Open Source Model Summary weist darauf hin, dass im Jahr 2026 der verstärkte Wettbewerb in Open-Source-Modellen zu Lizenzdifferenzierung führen wird: Google und Meta werden auf ARPACHE 2.0 umsteigen, während Zhipus GLM-5.3 von MIT auf eine benutzerdefinierte Lizenz umsteigen wird, was Inferenz- oder Feinabstimmungsdienste von Dienstleistern mit Jahresumsätzen von über 10 Milliarden Dollar erfordert, um Z.AI Sicherheitsüberprüfung zu bestehen, und Unsicherheit durch nicht näher bezeichnete Parteien wird Unsicherheit mit sich bringen.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News Ausgabe 24: Motip-3, GLM-5.3, Hy4-Vorschau und Trends in der Open-Source-Modelllizenzierung - Aioga KI-News","description":"Diese Ausgabe der Open Source Model Summary weist darauf hin, dass im Jahr 2026 der verstärkte Wettbewerb in Open-Source-Modellen zu Lizenzdifferenzierung führen wird: Google und M...","url":"https://www.aioga.com/de/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:12.104Z"},"pt-BR":{"title":"Open Source News Edição 24: Motip-3, GLM-5.3, prévia do Hy4 e tendências em licenciamento de modelos de código aberto","summary":"Esta edição do Resumo do Modelo de Código Aberto aponta que, em 2026, a concorrência intensificada em modelos de código aberto levará à diferenciação de licenças: Google e Meta migrarão para o ARPACHE 2.0, enquanto o GLM-5.3 da Zhipu mudará da licença do MIT para uma licença personalizada, exigindo inferência ou ajuste fino de provedores de serviços com receitas anuais superiores a 10 bilhões de dólares para passar Z.AI revisão de segurança, e a incerteza de partes não especificadas trará incerteza.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News Edição 24: Motip-3, GLM-5.3, prévia do Hy4 e tendências em licenciamento de modelos de código aberto - Aioga Notícias de IA","description":"Esta edição do Resumo do Modelo de Código Aberto aponta que, em 2026, a concorrência intensificada em modelos de código aberto levará à diferenciação de licenças: Google e Meta mig...","url":"https://www.aioga.com/pt-BR/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:21.547Z"},"ru":{"title":"Open Source News выпуск 24: Motip-3, GLM-5.3, Hy4-preview и тенденции лицензирования открытых моделей","summary":"В этом выпуске Open Source Model Summary отмечается, что в 2026 году усиленная конкуренция в открытых моделях приведёт к дифференциации лицензий: Google и Meta перейдут на ARPACHE 2.0, а GLM-5.3 от Zhipu перейдёт с MIT на пользовательскую лицензию, требуя вывода или тонкой настройки поставщиков услуг с годовой доходом свыше 10 миллиардов долларов для прохождения Z.AI проверке безопасности, а неопределённость со стороны неуказанных сторон вызовет неопределённость.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News выпуск 24: Motip-3, GLM-5.3, Hy4-preview и тенденции лицензирования открытых моделей - Aioga Новости ИИ","description":"В этом выпуске Open Source Model Summary отмечается, что в 2026 году усиленная конкуренция в открытых моделях приведёт к дифференциации лицензий: Google и Meta перейдут на ARPACHE...","url":"https://www.aioga.com/ru/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:21.774Z"},"ar":{"title":"أخبار المصدر المفتوح رقم 24: Motip-3، GLM-5.3، معاينة Hy4، واتجاهات ترخيص نماذج المصدر المفتوح","summary":"تشير هذه النسخة من ملخص النموذج مفتوح المصدر إلى أنه في عام 2026، سيؤدي التنافس المتزايد في نماذج المصدر المفتوح إلى تمييز في الترخيص: حيث ستنتقل جوجل وميتا إلى ARPACHE 2.0، بينما ستنتقل GLM-5.3 من شركة Zhipu من MIT إلى ترخيص مخصص، مما يتطلب الاستنتاج أو ضبط مزودي الخدمة الذين تتجاوز إيراداتهم السنوية 10 مليارات دولار لتجاوز مراجعة الأمان Z.AI، كما أن عدم اليقين من أطراف غير محددة سيجلب عدم اليقين.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"أخبار المصدر المفتوح رقم 24: Motip-3، GLM-5.3، معاينة Hy4، واتجاهات ترخيص نماذج المصدر المفتوح - Aioga أخبار الذكاء الاصطناعي","description":"تشير هذه النسخة من ملخص النموذج مفتوح المصدر إلى أنه في عام 2026، سيؤدي التنافس المتزايد في نماذج المصدر المفتوح إلى تمييز في الترخيص: حيث ستنتقل جوجل وميتا إلى ARPACHE 2.0، بينما...","url":"https://www.aioga.com/ar/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:31.460Z"},"hi":{"title":"ओपन सोर्स न्यूज़ अंक 24: Motip-3, GLM-5.3, Hy4-पूर्वावलोकन, और ओपन सोर्स मॉडल लाइसेंसिंग में रुझान","summary":"ओपन सोर्स मॉडल सारांश का यह अंक बताता है कि 2026 में, ओपन-सोर्स मॉडल में तीव्र प्रतिस्पर्धा से लाइसेंस भेदभाव होगा: Google और Meta ARPACHE 2.0 पर स्विच करेंगे, जबकि Zhipu का GLM-5.3 MIT से कस्टम लाइसेंस पर स्विच करेगा, जिसके लिए सुरक्षा समीक्षा पास करने के लिए $10 बिलियन से अधिक वार्षिक राजस्व वाले अनुमान या फाइन-ट्यूनिंग सेवा प्रदाताओं Z.AI आवश्यकता होगी, और अनिर्दिष्ट पार्टियों से अनिश्चितता अनिश्चितता लाएगी।","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"ओपन सोर्स न्यूज़ अंक 24: Motip-3, GLM-5.3, Hy4-पूर्वावलोकन, और ओपन सोर्स मॉडल लाइसेंसिंग में रुझान - Aioga AI समाचार","description":"ओपन सोर्स मॉडल सारांश का यह अंक बताता है कि 2026 में, ओपन-सोर्स मॉडल में तीव्र प्रतिस्पर्धा से लाइसेंस भेदभाव होगा: Google और Meta ARPACHE 2.0 पर स्विच करेंगे, जबकि Zhipu का GLM-5....","url":"https://www.aioga.com/hi/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:31.517Z"},"it":{"title":"Open Source News Numero 24: Motip-3, GLM-5.3, Hy4-preview e tendenze nel licenzing di modelli open source","summary":"Questo numero del Riepilogo del Modello Open Source sottolinea che nel 2026, una competizione intensificata nei modelli open-source porterà a differenziazioni sulle licenze: Google e Meta passeranno ad ARPACHE 2.0, mentre il GLM-5.3 di Zhipu passerà da una licenza MIT a una licenza personalizzata, richiedendo l'inferenza o il fine-tuning dei fornitori di servizi con ricavi annui superiori a 10 miliardi di dollari per superare Z.AI revisione della sicurezza, e l'incertezza da parte di parti non specificate porterà incertezza.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News Numero 24: Motip-3, GLM-5.3, Hy4-preview e tendenze nel licenzing di modelli open source - Aioga Notizie IA","description":"Questo numero del Riepilogo del Modello Open Source sottolinea che nel 2026, una competizione intensificata nei modelli open-source porterà a differenziazioni sulle licenze: Google...","url":"https://www.aioga.com/it/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:40.938Z"},"nl":{"title":"Open Source Nieuws Nummer 24: Motip-3, GLM-5.3, Hy4-preview en trends in Open Source Model Licensing","summary":"Deze editie van de Open Source Model Summary wijst erop dat in 2026 de intensieve concurrentie in open-source modellen zal leiden tot licentiedifferentiatie: Google en Meta zullen overstappen op ARPACHE 2.0, terwijl Zhipu's GLM-5.3 zal overstappen van MIT naar een aangepaste licentie, waardoor inferentie of verfijning van dienstverleners met een jaarlijkse omzet van meer dan $10 miljard moet slagen voor Z.AI beveiligingsbeoordeling, en onzekerheid van niet-gespecificeerde partijen zal zorgen voor onzekerheid.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source Nieuws Nummer 24: Motip-3, GLM-5.3, Hy4-preview en trends in Open Source Model Licensing - Aioga AI-nieuws","description":"Deze editie van de Open Source Model Summary wijst erop dat in 2026 de intensieve concurrentie in open-source modellen zal leiden tot licentiedifferentiatie: Google en Meta zullen...","url":"https://www.aioga.com/nl/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:40.865Z"},"tr":{"title":"Açık Kaynak Haber Sayı 24: Motip-3, GLM-5.3, Hy4-önizleme ve Açık Kaynak Model Lisanslama Trendleri","summary":"Açık Kaynak Model Özeti'nin bu sayısı, 2026'da açık kaynak modellerdeki artan rekabetin lisans farklılaşmasına yol açacağını vurguluyor: Google ve Meta ARPACHE 2.0'a geçecek, Zhipu'nun GLM-5.3'ü ise MIT'den özel lisansa geçecek; bu da yıllık geliri 10 milyar doları aşan hizmet sağlayıcılarının güvenlik incelemesinden geçmesi için çıkarım veya ince ayar yapması Z.AI ve belirsiz taraflardan gelen belirsizlik belirsizlik getirecek.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Açık Kaynak Haber Sayı 24: Motip-3, GLM-5.3, Hy4-önizleme ve Açık Kaynak Model Lisanslama Trendleri - Aioga AI Haberleri","description":"Açık Kaynak Model Özeti'nin bu sayısı, 2026'da açık kaynak modellerdeki artan rekabetin lisans farklılaşmasına yol açacağını vurguluyor: Google ve Meta ARPACHE 2.0'a geçecek, Zhipu...","url":"https://www.aioga.com/tr/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:50.366Z"},"vi":{"title":"Tin tức Mã nguồn Mở số 24: Motip-3, GLM-5.3, Hy4-preview và Xu hướng trong Cấp phép Mô hình Mã nguồn Mở","summary":"Số này của Tóm tắt Mô hình Mã nguồn Mở chỉ ra rằng vào năm 2026, sự cạnh tranh gay gắt trong các mô hình mã nguồn mở sẽ dẫn đến sự phân biệt giấy phép: Google và Meta sẽ chuyển sang ARPACHE 2.0, trong khi GLM-5.3 của Zhipu sẽ chuyển từ MIT sang giấy phép tùy chỉnh, đòi hỏi phải suy luận hoặc tinh chỉnh các nhà cung cấp dịch vụ có doanh thu hàng năm vượt quá 10 tỷ đô la để vượt qua Z.AI đánh giá bảo mật, và sự không chắc chắn từ các bên chưa xác định sẽ mang lại sự không chắc chắn.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Tin tức Mã nguồn Mở số 24: Motip-3, GLM-5.3, Hy4-preview và Xu hướng trong Cấp phép Mô hình Mã nguồn Mở - Tin tức AI Aioga","description":"Số này của Tóm tắt Mô hình Mã nguồn Mở chỉ ra rằng vào năm 2026, sự cạnh tranh gay gắt trong các mô hình mã nguồn mở sẽ dẫn đến sự phân biệt giấy phép: Google và Meta sẽ chuyển san...","url":"https://www.aioga.com/vi/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:24:50.295Z"},"id":{"title":"Berita Open Source Edisi 24: Motip-3, GLM-5.3, Hy4-preview, dan Tren dalam Lisensi Model Open Source","summary":"Edisi Ringkasan Model Open Source ini menunjukkan bahwa pada tahun 2026, persaingan yang semakin ketat dalam model sumber terbuka akan menyebabkan diferensiasi lisensi: Google dan Meta akan beralih ke ARPACHE 2.0, sementara GLM-5.3 milik Zhipu akan beralih dari MIT ke lisensi khusus, yang mengharuskan inferensi atau penyempurnaan penyedia layanan dengan pendapatan tahunan melebihi $10 miliar untuk lolos tinjauan keamanan Z.AI, dan ketidakpastian dari pihak yang tidak disebutkan akan menimbulkan ketidakpastian.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Berita Open Source Edisi 24: Motip-3, GLM-5.3, Hy4-preview, dan Tren dalam Lisensi Model Open Source - Berita AI Aioga","description":"Edisi Ringkasan Model Open Source ini menunjukkan bahwa pada tahun 2026, persaingan yang semakin ketat dalam model sumber terbuka akan menyebabkan diferensiasi lisensi: Google dan...","url":"https://www.aioga.com/id/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:25:00.004Z"},"th":{"title":"ข่าวโอเพนซอร์ส ฉบับที่ 24: Motip-3, GLM-5.3, Hy4-preview และแนวโน้มในการอนุญาตใช้โมเดลโอเพนซอร์ส","summary":"ฉบับนี้ของ Open Source Model Summary ชี้ให้เห็นว่าในปี 2026 การแข่งขันที่เข้มข้นขึ้นในโมเดลโอเพนซอร์สจะนําไปสู่ความแตกต่างของใบอนุญาต: Google และ Meta จะเปลี่ยนไปใช้ ARPACHE 2.0 ขณะที่ GLM-5.3 ของ Zhipu จะเปลี่ยนจาก MIT เป็นใบอนุญาตแบบกําหนดเอง ซึ่งต้องมีการอนุมานหรือปรับแต่งผู้ให้บริการที่มีรายได้ต่อปีเกิน 10 พันล้านดอลลาร์เพื่อ Z.AI ผ่านการตรวจสอบความปลอดภัย และความไม่แน่นอนจากฝ่ายที่ไม่ระบุจะนํามาซึ่งความไม่แน่นอน","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"ข่าวโอเพนซอร์ส ฉบับที่ 24: Motip-3, GLM-5.3, Hy4-preview และแนวโน้มในการอนุญาตใช้โมเดลโอเพนซอร์ส - ข่าว AI Aioga","description":"ฉบับนี้ของ Open Source Model Summary ชี้ให้เห็นว่าในปี 2026 การแข่งขันที่เข้มข้นขึ้นในโมเดลโอเพนซอร์สจะนําไปสู่ความแตกต่างของใบอนุญาต: Google และ Meta จะเปลี่ยนไปใช้ ARPACHE 2.0 ขณ...","url":"https://www.aioga.com/th/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:25:00.165Z"},"pl":{"title":"Open Source News Numer 24: Motip-3, GLM-5.3, Hy4-preview i trendy w licencjonowaniu modeli open source","summary":"W tym wydaniu Open Source Model Summary wskazuje, że w 2026 roku zaostrzona konkurencja w modelach open source doprowadzi do różnicowania licencji: Google i Meta przejdą na ARPACHE 2.0, podczas gdy GLM-5.3 Zhipu przejdzie z MIT na niestandardową licencję, wymagając od dostawców wnioskowania lub dopracowywania usług o rocznych przychodach przekraczających 10 miliardów dolarów Z.AI przeglądzie bezpieczeństwa, a niepewność ze strony nieokreślonych stron przyniesie niepewność.","category":"行业动态","source":"Nathan Lambert：Interconnects（RSS）","aggregationSource":"Nathan Lambert：Interconnects（RSS）","pageTitle":"Open Source News Numer 24: Motip-3, GLM-5.3, Hy4-preview i trendy w licencjonowaniu modeli open source - Aioga Wiadomości AI","description":"W tym wydaniu Open Source Model Summary wskazuje, że w 2026 roku zaostrzona konkurencja w modelach open source doprowadzi do różnicowania licencji: Google i Meta przejdą na ARPACHE...","url":"https://www.aioga.com/pl/news/cmtssoxb003vvrokaz4fxgjpl/","contentTranslated":true,"sourceHash":"d66f2ba0e52d405d","translatedAt":"2026-09-08T15:25:09.544Z"}},"evidenceTier":"verified-news","reviewStatus":"automated-ingest","indexable":true,"editorialCover":"/page-visuals/topic-timeline.png"}}