开放式AI的地缘政治今年秋天又在加速发展，我们预计所有AI系统(开放和封闭)的风险会迅速上升。 我们上周看到许多新闻，显示美国和中国政府都在对开放模式进行控制。
— MOFCOM，中华人民共和国商务部，发布一项政策提案(目前还不是法律)，要求前沿AI模型和一些其他技术获得政府预先批准后才能在国际上共享。 我们发现对此的反应过于夸张 — — 该文件指出，开源技术将被设置为自动审批 — — 但地缘政治圈中许多人认为这很重要，因为它提出了一个监管框架，可以在未来用于监管中国AI产业。
我们相信的另一些人说，这个提议可能更多是防止诸如向西方公司出售中国创业者马努斯之类的问题，而不是涉及前沿AI模型。 无论哪种方式，控制车轮都在AI周围移动，开放量级的车型也不能不受任何影响。
- 国土安全部(国土安全部)发布对DeepSeek下达命令，限制DeepSeek作为公司本身，限制DeepSeek作为技术平台的型号成为联邦民间采购的供应商。 这与经过精细调整的DeepSeek模型或模型有何关系尚不清楚(例如，Nvidia的Nemotron模型在培训食谱中长期严重依赖DeepSeek模型 — — 它们将被禁用吗？)，但这是一个预期的行动，并强化了对国内开放模型生态系统的需求。
中国的其他模型制造者，如GLM的制造者Zhipu AI，已经被列入商业/BIS下的实体名单，这限制了他们获得美国技术的机会。
DeepSeek跟随上述消息，公开发布其内部推论堆栈DeepInfer，这很可能比现有的开源工具(如VLLM/SGLang)为他们的模型效率要高得多。 估计表明DeepSeek在这个推论堆上出售信物时总比值高达90%. 众所周知，顶级AI实验室内部有更高效的推论工具，这是我们第一次看到领先的实验室共享这些工具。 推论公司会很快采用这些工具，虽然不是一个开放式模型或数据集，但我们认为这一发布是过去几个月中更大的开源 AI文物之一。
在写这个时，Kimi K3.1传闻很快成为发布，与之相关的元数据开始出现在公共图书馆。 令人兴奋的时光！
您可以通过这封邮件查看模型，以及以前的所有问题，在我们的Artifacts Hub上。 其中包括评注和其他生态系统元数据。
我们的选手
-MiMo-V2.6-Pro-RL by XiaomimiMo:Mimo的更新，它重新使用与MiMo V2.5相同的基地，但在RL训练上却使用了更多的资源。 他们还发布了一份深入的技术报告，以及一些使用过的环境，这在这种规模上是罕见的。 这个食谱的最终型号V2.6-Pro是目前人工分析排名的顶级开放型号，规模化的RL运行伴随着一个有趣的实时训练仪表板。 我们数月来一直是小米的粉丝
- DeepSeek-V4.1-Flash-1017 by deepseek-ai：自最新人工兽以来，鲸鱼没有发布 1，而是有两个新的模型迭代，其新的尺寸(552B-A16B)也支持图像输入。 该型号以小价廉的性能获得了许多(当之无愧)的关注。 它还携带麻省理工学院的许可证，延续了DeepSeek自己确立的趋势。 在许可证似乎更加公开和排斥的情况下，这一点是显著的。 下面是DeepSWE上的性能与价格的Pareto，这是当今最喜爱的代理软件工程基准测试，DeepSeek占主导地位。
- Cloudflare的clef：决定模型都是推出在启动Typesafe之后的愤怒。 简言之，它们是一般分类模型，其概率输出的符号很少。 它的实用性导致了快速的采用，以及许多开放模型的微调，如克莱夫。
- NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4 by nvidia：尼莫特龙家族的中型模型得到了它期待已久的更新。 它的体积很强，可以在家里运行(因为你拥有一台牛肉般的笔记本电脑或RTX Spark). 同往常一样，NVIDIA 发布是绝大多数的数据和深入的技术报告。
- Beam-1.0-501B-A23B通过反射：由Reflection 期待已久的开放模式，使其在许可(Apache 2.0)下成为最强开放的美国模式发布。 Beam虽然不是现有最聪明的开放型号，但具有非常令人印象深刻的象征效率(次于下方的GPT-6-Astra). 这标志着训练后堆积的发达。
模型
一般用途
- Ternary-Bonsai-227B-gguf by prism-ml：一个Ternary版本的Qwen3.8-27B用于端侧部署。
- Ornith-ai的Ornith-1.5-397B：一系列Qwen3.5和Gemma 4微调，这些微调因在相应(强)基线上的强效表现而在当地AI社区引起关注。
- Ling-3.0-flash-VL-fp4 by universityAI：由Ant Ling(在Ant Group内部)制作的124B-A5.5B型号，该型号继续持续地喷出型号。
- Muse-Spark-Open-816B-A42B by meta-models:META有发布的开放版本 Muse，我们更详细地写了这个版本！
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
The geopolitics of open-weight AI are accelerating again this fall, as we would expect with the risks of all AI systems (open and closed) rising rapidly. We’ve seen many news items in the last week showing both the U.S. and Chinese governments positioning themselves for some controls on open models.
- MOFCOM, the Ministry of Commerce of the People’s Republic of China, released a policy proposal (it’s not yet a law) that would require frontier AI models and some other technologies to get pre-approval from the government before they can be shared internationally. We found the reaction to this to be overblown – as the document states that open-source technologies will be set up for automatic approvals – but many in geopolitical circles say this is a big deal because it proposes a regulatory framework that could be used in the future to regulate China’s AI industry.
Others we trust on this issue say this proposal could be more about preventing issues like the sale of Chinese startup Manus to Western companies, rather than being about frontier AI models. Either way, the wheels of control are moving around AI, and open-weight models are not free from any influence.
- The Department of Homeland Security (DHS) released an order on DeepSeek, which restricts DeepSeek itself, as a company, and DeepSeek’s models, as a technology platform, from being a vendor in federal civilian procurement. It is not clear how this will relate to finetuned models or models trained on the outputs of DeepSeek (e.g., Nvidia’s Nemotron models have long relied heavily on DeepSeek models in their training recipes – will they be banned?), but this is an expected action and reinforces the need for a domestic open model ecosystem.
Other model makers in China, such as Zhipu AI, makers of GLM, are already on the Entity List under Commerce/BIS, which restricts their access to American technology.
DeepSeek followed the above news up by releasing their internal inference stack publicly, DeepInfer, which is likely far more efficient than existing open-source tools like vLLM/SGLang for their models. Estimates point to DeepSeek making up to 90% gross margin when selling tokens on this inference stack. It is well-known that the top AI labs have much more inference-efficient tools internally, and this is the first time we’ve seen a leading lab share them. Inference companies will be quick to adopt these tools, which will drive prices for DeepSeek’s models down across the ecosystem (or margins up). While not an open model or dataset, we consider this release to be one of the bigger open-source AI artifacts in the last few months.
As of writing this, Kimi K3.1 is rumored to be released very soon, with metadata associated with it starting to appear in public libraries. Exciting times!
You can view the models in this email, and all previous issues, on our Artifacts Hub. It includes commentary and other ecosystem metadata.
Our Picks
- MiMo-V2.6-Pro-RL by XiaomiMiMo: an update for Mimo, which reuses the same bases as MiMo V2.5, but used a lot more resources on RL training. They also release an in-depth tech report, as well as some of the used environments, which is rare at this scale. The final model from this recipe, V2.6-Pro, is the top open model on Artificial Analysis’s ranking today, and the scaled RL runs came with a fun live dashboard of training. We’ve been fans of Xiaomi for months now, and their efforts are bearing fruit.
- DeepSeek-V4.1-Flash-1017 by deepseek-ai: Since the latest Artifacts, the whale has not released one, but two new model iterations with a new size (552B-A16B) which also support image inputs. The model has gained a lot of (well-deserved) attention for its performance at a small price. It also carries the MIT license, continuing the trend established by DeepSeek themselves. This is notable in times where licenses seem to become more bespoke and exclusionary. Below is the Pareto of performance vs price on DeepSWE, a favorite agentic software engineering benchmark today, which DeepSeek dominates.
- clef by Cloudflare: Decision models are all the rage after being introduced by the startup Typesafe. In short, they are general classification models that output few tokens with their probabilities. Its usefulness has led to quick adoption and also a lot of open model finetunes such as Clef.
- NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4 by nvidia: The mid-sized model of the Nemotron family got its long awaited update. It is strong for its size and can be run at home (given you own a beefy laptop or RTX Spark). As usual, NVIDIA releases the vast majority of the data and an in-depth tech report.
- Beam-1.0-501B-A23B by reflection: The long awaited open model by Reflection, making it the strongest open American model released under a permissive license (Apache 2.0). While not the smartest open model available, Beam is very impressively token efficient (second to only GPT-6-Astra below). This is a sign of a well-developed post-training stack.
Models
General Purpose
- Ternary-Bonsai-2-27B-gguf by prism-ml: A ternary version of Qwen3.8-27B for on-device deployments.
- Ornith-1.5-397B by ornith-ai: A series of Qwen3.5 and Gemma 4 finetunes that gained attention in the local AI community for its strong performance over the respective (strong) baselines.
- Ling-3.0-flash-VL-fp4 by inclusionAI: A 124B-A5.5B model by Ant Ling (within Ant Group), which continues to churn out models consistently.
- Muse-Spark-Open-816B-A42B by meta-models: META has released their open version of Muse, which we have written about in more detail!
Auto-fetched and translated by AI Atlas. All rights belong to the original author.