MiniMax M2.7模型登陆NVIDIA GPU加速端点，支持NemoClaw与OpenClaw

ME News Update, April 12 (UTC+8), MiniMax announced that its M2.7 model is now available on NVIDIA GPU-accelerated endpoints and supports tools including NemoClaw and OpenClaw. M2.7 is the latest version of the M2 series, designed to enhance complex applications such as agent workflows, reasoning, and machine learning research. It is a sparse mixture-of-experts model with a total of 230 billion parameters, 10 billion active parameters per token, an activation rate of approximately 4.3%, supporting a context length of 200k tokens, with 256 local experts, 8 experts activated per token, across 62 layers. To support deployment, NVIDIA has released the open-source reference stack NemoClaw to simplify the secure deployment of OpenClaw, which can be launched with a single click via the NVIDIA Brev cloud platform. Additionally, NVIDIA has collaborated with the open-source community to integrate high-performance kernels optimized for large-scale MoE models into frameworks such as vLLM and SGLang. (Source: InFoQ)