5月9日,蚂蚁百灵发布万亿级旗舰思考模型Ring-2.6-1T,专为复杂真实任务场景设计。模型创新引入可调节Reasoning Effort机制,支持high与xhigh两档推理强度。
在真实任务执行类评测中,PinchBench得分87.6,高于GPT-5.4x High、Gemini-3.1-Pro high与Claude-Opus-4.7x high;在高难推理任务上,ARC-agi-V2得分77.78,与Gemini-3.1-Pro high和Claude-Opus-4.7x high处于同一水位。
![]()
特别声明:以上内容(如有图片或视频亦包括在内)为自媒体平台“网易号”用户上传并发布,本平台仅提供信息存储服务。
Notice: The content above (including the pictures and videos if any) is uploaded and posted by a user of NetEase Hao, which is a social media platform and only provides information storage services.