Cognition 发布编码模型 SWE-2 Flash,基于 Inkling-Small 构建
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 opt
AI 导读
Cognition 发布快速编码模型 SWE-2 Flash,基于 thinkymachines 的 Inkling-Small 构建,总参数 276B、激活参数 12B。官方称其成本比 SWE-2 低 5 倍、速度快 3 倍,在主要评测上超过所有美国开源权重模型,包括规模大 3 倍的模型。该模型本月内对 Devin Desktop 和 Devin CLI 的 Pro、Max、Teams 用户免费,权重将通过合作项目定向开放。
正文
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. https://t.co/Se49qXj9sd