精选理由
GLM-5.3-Flash模型发布，专为高效编码和长周期智能体任务设计，比传统模型更高效。
GLM-5.3-Flash模型推出，融合稀疏与线性注意力架构，提升长上下文能力，降低计算负担。多供应商预计本周部署。
原文 · OpenRouter
GLM-5.3-Flash is built for efficient coding and long-horizon agent tasks. Its hybrid sparse and lin...
GLM-5.3-Flash is built for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture preserves accurate long-context capabilities while reducing compute overhead. We expect many providers to onboard this model throughout the week. 💬 2 🔄 0 ❤️ 38 👀 3498 📊 5 ⚡