Cognition· @cognition · X·· 3 小时前AI 评分60
AI 导读
Cognition 推出 SWE-2 Flash，该模型基于 Inkling-Small 构建（拥有 276B 总参数和 12B 激活参数）。结合高度优化的推理栈，其成本比 SWE-2 降低 5 倍，速度提升 3 倍。
正文 · 原文
SWE-2 Flash is based on @thinkymachines Inkling-Small, an American model with 276B total and 12B active parameters. The small size, combined with our highly optimized inference stack allows us to offer it at a 5x lower cost and 3x higher speed than SWE-2.
来源：Cognition · x.com