Thinking Machines 发布开源模型 Inkling 2 Mini
26B 参数但活跃只有 5.4B,MacBook 就能跑,Apache 2.0 可商用,还带 1M 上下文。想在自己设备上部署开源模型、或对比同量级 MoE 效率的开发者,可以直接下权重试试。
Thinking Machines 发布了开源权重模型 Inkling 2 Mini,在 Artificial Analysis Intelligence Index 上拿到 20 分,处于智能水平与活跃参数量的 Pareto 前沿。该模型总参数 26B,采用 MoE 架构,活跃参数仅 5.4B,规模小到可以在 MacBook 或 DGX Spark 上本地运行。
技术上它是一个 looped transformer:5.4B 活跃参数包含三次循环层前向传播,实际每个文本 token 只用到 3.1B 的独立权重,用重复计算换取低参数量下的表现。模型支持文本与图像输入,上下文窗口达 1M token,采用 Apache 2.0 许可证发布。
对开发者来说,这意味着一个可以在消费级硬件上跑、可商用、还有 1M 上下文的多模态开源选项;在同等活跃参数量级的开源模型里,它的 Intelligence Index 得分给出了一个新参照点。
Thinking Machines has released Inkling 2 Mini at 20 on the Artificial Analysis Intelligence Index, an open weights looped transformer on the Pareto frontier for Intelligence vs. Active Parameters
Inkling 2 Mini is a 26B total, 5.4B active parameter mixture-of-experts model. The 5.4B active parameter count includes three passes through its looped layers; it uses 3.1B unique weights per text token, and it is small enough to run on a MacBook or DGX Spark.
It is released under an Apache 2.0 license, with text and image input and a 1M token context window.
Congratulations to @thinkymachines on the release! 🧵
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Inkling 2 Mini, Thinking Machines, open-source models, MoE.
Source attribution: curated from Artificial Analysis (@ArtificialAnlys). Cross-source corroboration: 6 independent sources have covered this development (including kimmonismus, Thinking Machines Lab, Merve Noyan), confirming industry-wide signal strength.