精选理由
DeepSeek Harness 推出 v0.1.0 - rc.8 更新，加了图像请求和多模态功能，比之前能处理更多内容的了。
DeepSeek Harness 发布 v0.1.0 - rc.8 更新，为自公开测试版以来的首次重要更新；本次更新包含 14 项变更，涉及多模态输入与子代理协作等方面；新增原生图像请求与混合文本图像输入功能，使代理能够处理截图；还通过 OCR 与像素分析实现‘视觉’功能，让原本仅支持文本的模型也能‘看见’；
原文 · Pandaily
DeepSeek Harness Ships First Major Post-Beta Update: 14 Changes Bring Multimodal and Subagent Upgrades
Last night DeepSeek Harness released v0.1.0-rc.8, its first significant update since public beta, with 14 changes covering multimodal input, subagent collaboration, terminal experience, tool calling, and developer support. New native image requests and mixed text-image input let agents handle screenshots, while a tool-layer 'vision' via OCR and pixel analysis gives even text-only models a way to 'see'.